diff --git a/Android.bp b/Android.bp
index f14eb21..59d5995 100644
--- a/Android.bp
+++ b/Android.bp
@@ -412,6 +412,8 @@
         "src/core/Utils.cpp",
         "src/core/Validate.cpp",
         "src/core/Version.cpp",
+        "src/core/helpers/SoftmaxHelpers.cpp",
+        "src/core/helpers/WindowHelpers.cpp",
         "src/core/utils/ScaleUtils.cpp",
         "src/core/utils/helpers/fft.cpp",
         "src/core/utils/helpers/tensor_transform.cpp",
diff --git a/SConscript b/SConscript
index b915557..1b03629 100644
--- a/SConscript
+++ b/SConscript
@@ -178,6 +178,7 @@
 core_files = Glob('src/core/*.cpp')
 core_files += Glob('src/core/CPP/*.cpp')
 core_files += Glob('src/core/CPP/kernels/*.cpp')
+core_files += Glob('src/core/helpers/*.cpp')
 core_files += Glob('src/core/utils/*.cpp')
 core_files += Glob('src/core/utils/helpers/*.cpp')
 core_files += Glob('src/core/utils/io/*.cpp')
@@ -228,11 +229,10 @@
     # build winograd/depthwise sources for either v7a / v8a
     core_files += Glob('src/core/NEON/kernels/convolution/*/*.cpp')
     core_files += Glob('src/core/NEON/kernels/convolution/winograd/*/*.cpp')
-    arm_compute_env.Append(CPPPATH = ["arm_compute/core/NEON/kernels/convolution/common/",
-                                      "arm_compute/core/NEON/kernels/convolution/winograd/",
-                                      "arm_compute/core/NEON/kernels/convolution/depthwise/",
-                                      "src/core/NEON/kernels/assembly/",
+    arm_compute_env.Append(CPPPATH = ["src/core/NEON/kernels/convolution/common/",
                                       "src/core/NEON/kernels/convolution/winograd/",
+                                      "src/core/NEON/kernels/convolution/depthwise/",
+                                      "src/core/NEON/kernels/assembly/",
                                       "arm_compute/core/NEON/kernels/assembly/"])
 
     graph_files += Glob('src/graph/backends/NEON/*.cpp')
diff --git a/arm_compute/core/CL/ICLKernel.h b/arm_compute/core/CL/ICLKernel.h
index d4990a1..a24cd8c 100644
--- a/arm_compute/core/CL/ICLKernel.h
+++ b/arm_compute/core/CL/ICLKernel.h
@@ -29,6 +29,7 @@
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/GPUTarget.h"
 #include "arm_compute/core/IKernel.h"
+#include "arm_compute/core/Validate.h"
 #include "arm_compute/core/experimental/Types.h"
 
 #include <string>
diff --git a/arm_compute/core/GPUTarget.h b/arm_compute/core/GPUTarget.h
index 06025ca..b8143f8 100644
--- a/arm_compute/core/GPUTarget.h
+++ b/arm_compute/core/GPUTarget.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_GPUTARGET_H
 #define ARM_COMPUTE_GPUTARGET_H
 
-#include "arm_compute/core/Helpers.h"
+#include "support/Traits.h"
 
 #include <string>
 
diff --git a/arm_compute/core/Helpers.h b/arm_compute/core/Helpers.h
index 90dd608..5a8d6ef 100644
--- a/arm_compute/core/Helpers.h
+++ b/arm_compute/core/Helpers.h
@@ -24,23 +24,17 @@
 #ifndef ARM_COMPUTE_HELPERS_H
 #define ARM_COMPUTE_HELPERS_H
 
-#include "arm_compute/core/Coordinates.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/IAccessWindow.h"
-#include "arm_compute/core/Steps.h"
-#include "arm_compute/core/Strides.h"
-#include "arm_compute/core/TensorShape.h"
+#include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
+#include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
-#include "support/MemorySupport.h"
 
 #include <array>
 #include <cstddef>
 #include <cstdint>
-#include <memory>
 #include <tuple>
-#include <type_traits>
-#include <utility>
 
 namespace arm_compute
 {
@@ -48,307 +42,6 @@
 class ITensor;
 class ITensorInfo;
 
-/** Disable bitwise operations by default */
-template <typename T>
-struct enable_bitwise_ops
-{
-    static constexpr bool value = false; /**< Disabled */
-};
-
-#ifndef DOXYGEN_SKIP_THIS
-template <typename T>
-typename std::enable_if<enable_bitwise_ops<T>::value, T>::type operator&(T lhs, T rhs)
-{
-    using underlying_type = typename std::underlying_type<T>::type;
-    return static_cast<T>(static_cast<underlying_type>(lhs) & static_cast<underlying_type>(rhs));
-}
-#endif /* DOXYGEN_SKIP_THIS */
-
-/** Helper function to create and return a unique_ptr pointed to a CL/GLES kernel object
- *  It also calls the kernel's configuration.
- *
- * @param[in] args All the arguments that need pass to kernel's configuration.
- *
- * @return A unique pointer pointed to a CL/GLES kernel object
- */
-template <typename Kernel, typename... T>
-std::unique_ptr<Kernel> create_configure_kernel(T &&... args)
-{
-    std::unique_ptr<Kernel> k = arm_compute::support::cpp14::make_unique<Kernel>();
-    k->configure(std::forward<T>(args)...);
-    return k;
-}
-
-/** Helper function to create and return a unique_ptr pointed to a CL/GLES kernel object
- *
- * @return A unique pointer pointed to a Kernel kernel object
- */
-template <typename Kernel>
-std::unique_ptr<Kernel> create_kernel()
-{
-    std::unique_ptr<Kernel> k = arm_compute::support::cpp14::make_unique<Kernel>();
-    return k;
-}
-
-namespace traits
-{
-/** Check if a type T is contained in a tuple Tuple of types */
-template <typename T, typename Tuple>
-struct is_contained;
-
-template <typename T>
-struct is_contained<T, std::tuple<>> : std::false_type
-{
-};
-
-template <typename T, typename... Ts>
-struct is_contained<T, std::tuple<T, Ts...>> : std::true_type
-{
-};
-
-template <typename T, typename U, typename... Ts>
-struct is_contained<T, std::tuple<U, Ts...>> : is_contained<T, std::tuple<Ts...>>
-{
-};
-}
-
-/** Computes bilinear interpolation using the pointer to the top-left pixel and the pixel's distance between
- * the real coordinates and the smallest following integer coordinates. Input must be in single channel format.
- *
- * @param[in] pixel_ptr Pointer to the top-left pixel value of a single channel input.
- * @param[in] stride    Stride to access the bottom-left and bottom-right pixel values
- * @param[in] dx        Pixel's distance between the X real coordinate and the smallest X following integer
- * @param[in] dy        Pixel's distance between the Y real coordinate and the smallest Y following integer
- *
- * @note dx and dy must be in the range [0, 1.0]
- *
- * @return The bilinear interpolated pixel value
- */
-template <typename T>
-inline T delta_bilinear_c1(const T *pixel_ptr, size_t stride, float dx, float dy)
-{
-    ARM_COMPUTE_ERROR_ON(pixel_ptr == nullptr);
-
-    const float dx1 = 1.0f - dx;
-    const float dy1 = 1.0f - dy;
-
-    const T a00 = *pixel_ptr;
-    const T a01 = *(pixel_ptr + 1);
-    const T a10 = *(pixel_ptr + stride);
-    const T a11 = *(pixel_ptr + stride + 1);
-
-    const float w1 = dx1 * dy1;
-    const float w2 = dx * dy1;
-    const float w3 = dx1 * dy;
-    const float w4 = dx * dy;
-
-    return static_cast<T>(a00 * w1 + a01 * w2 + a10 * w3 + a11 * w4);
-}
-
-/** Computes bilinear interpolation for quantized input and output, using the pointer to the top-left pixel and the pixel's distance between
- * the real coordinates and the smallest following integer coordinates. Input must be QASYMM8 and in single channel format.
- *
- * @param[in] pixel_ptr Pointer to the top-left pixel value of a single channel input.
- * @param[in] stride    Stride to access the bottom-left and bottom-right pixel values
- * @param[in] dx        Pixel's distance between the X real coordinate and the smallest X following integer
- * @param[in] dy        Pixel's distance between the Y real coordinate and the smallest Y following integer
- * @param[in] iq_info   Input QuantizationInfo
- * @param[in] oq_info   Output QuantizationInfo
- *
- * @note dx and dy must be in the range [0, 1.0]
- *
- * @return The bilinear interpolated pixel value
- */
-inline uint8_t delta_bilinear_c1_quantized(const uint8_t *pixel_ptr, size_t stride, float dx, float dy, UniformQuantizationInfo iq_info, UniformQuantizationInfo oq_info)
-{
-    ARM_COMPUTE_ERROR_ON(pixel_ptr == nullptr);
-
-    const float dx1 = 1.0f - dx;
-    const float dy1 = 1.0f - dy;
-
-    const float a00 = dequantize_qasymm8(*pixel_ptr, iq_info);
-    const float a01 = dequantize_qasymm8(*(pixel_ptr + 1), iq_info);
-    const float a10 = dequantize_qasymm8(*(pixel_ptr + stride), iq_info);
-    const float a11 = dequantize_qasymm8(*(pixel_ptr + stride + 1), iq_info);
-
-    const float w1  = dx1 * dy1;
-    const float w2  = dx * dy1;
-    const float w3  = dx1 * dy;
-    const float w4  = dx * dy;
-    float       res = a00 * w1 + a01 * w2 + a10 * w3 + a11 * w4;
-    return static_cast<uint8_t>(quantize_qasymm8(res, oq_info));
-}
-
-/** Computes bilinear interpolation for quantized input and output, using the pointer to the top-left pixel and the pixel's distance between
- * the real coordinates and the smallest following integer coordinates. Input must be QASYMM8_SIGNED and in single channel format.
- *
- * @param[in] pixel_ptr Pointer to the top-left pixel value of a single channel input.
- * @param[in] stride    Stride to access the bottom-left and bottom-right pixel values
- * @param[in] dx        Pixel's distance between the X real coordinate and the smallest X following integer
- * @param[in] dy        Pixel's distance between the Y real coordinate and the smallest Y following integer
- * @param[in] iq_info   Input QuantizationInfo
- * @param[in] oq_info   Output QuantizationInfo
- *
- * @note dx and dy must be in the range [0, 1.0]
- *
- * @return The bilinear interpolated pixel value
- */
-inline int8_t delta_bilinear_c1_quantized(const int8_t *pixel_ptr, size_t stride, float dx, float dy, UniformQuantizationInfo iq_info, UniformQuantizationInfo oq_info)
-{
-    ARM_COMPUTE_ERROR_ON(pixel_ptr == nullptr);
-
-    const float dx1 = 1.0f - dx;
-    const float dy1 = 1.0f - dy;
-
-    const float a00 = dequantize_qasymm8_signed(*pixel_ptr, iq_info);
-    const float a01 = dequantize_qasymm8_signed(*(pixel_ptr + 1), iq_info);
-    const float a10 = dequantize_qasymm8_signed(*(pixel_ptr + stride), iq_info);
-    const float a11 = dequantize_qasymm8_signed(*(pixel_ptr + stride + 1), iq_info);
-
-    const float w1  = dx1 * dy1;
-    const float w2  = dx * dy1;
-    const float w3  = dx1 * dy;
-    const float w4  = dx * dy;
-    float       res = a00 * w1 + a01 * w2 + a10 * w3 + a11 * w4;
-    return static_cast<int8_t>(quantize_qasymm8_signed(res, oq_info));
-}
-
-/** Computes linear interpolation using the pointer to the top pixel and the pixel's distance between
- * the real coordinates and the smallest following integer coordinates. Input must be in single channel format.
- *
- * @param[in] pixel_ptr Pointer to the top pixel value of a single channel input.
- * @param[in] stride    Stride to access the bottom pixel value
- * @param[in] dy        Pixel's distance between the Y real coordinate and the smallest Y following integer
- *
- * @note dy must be in the range [0, 1.0]
- *
- * @return The linear interpolated pixel value
- */
-template <typename T>
-inline T delta_linear_c1_y(const T *pixel_ptr, size_t stride, float dy)
-{
-    ARM_COMPUTE_ERROR_ON(pixel_ptr == nullptr);
-
-    const float dy1 = 1.0f - dy;
-
-    const T a00 = *pixel_ptr;
-    const T a10 = *(pixel_ptr + stride);
-
-    const float w1 = dy1;
-    const float w3 = dy;
-
-    return static_cast<T>(a00 * w1 + a10 * w3);
-}
-/** Computes linear interpolation using the pointer to the left pixel and the pixel's distance between
- * the real coordinates and the smallest following integer coordinates. Input must be in single channel format.
- *
- * @param[in] pixel_ptr Pointer to the left pixel value of a single channel input.
- * @param[in] dx        Pixel's distance between the X real coordinate and the smallest X following integer
- *
- * @note dx must be in the range [0, 1.0]
- *
- * @return The linear interpolated pixel value
- */
-template <typename T>
-inline T delta_linear_c1_x(const T *pixel_ptr, float dx)
-{
-    ARM_COMPUTE_ERROR_ON(pixel_ptr == nullptr);
-
-    const T a00 = *pixel_ptr;
-    const T a01 = *(pixel_ptr + 1);
-
-    const float dx1 = 1.0f - dx;
-
-    const float w1 = dx1;
-    const float w2 = dx;
-
-    return static_cast<T>(a00 * w1 + a01 * w2);
-}
-/** Return the pixel at (x,y) using bilinear interpolation.
- *
- * @warning Only works if the iterator was created with an IImage
- *
- * @param[in] first_pixel_ptr Pointer to the first pixel of a single channel input.
- * @param[in] stride          Stride in bytes of the image;
- * @param[in] x               X position of the wanted pixel
- * @param[in] y               Y position of the wanted pixel
- *
- * @return The pixel at (x, y) using bilinear interpolation.
- */
-template <typename T>
-inline T pixel_bilinear_c1(const T *first_pixel_ptr, size_t stride, float x, float y)
-{
-    ARM_COMPUTE_ERROR_ON(first_pixel_ptr == nullptr);
-
-    const int32_t xi = std::floor(x);
-    const int32_t yi = std::floor(y);
-
-    const float dx = x - xi;
-    const float dy = y - yi;
-
-    return delta_bilinear_c1(first_pixel_ptr + xi + yi * stride, stride, dx, dy);
-}
-
-/** Return the pixel at (x,y) using bilinear interpolation by clamping when out of borders. The image must be single channel input
- *
- * @warning Only works if the iterator was created with an IImage
- *
- * @param[in] first_pixel_ptr Pointer to the first pixel of a single channel image.
- * @param[in] stride          Stride in bytes of the image
- * @param[in] width           Width of the image
- * @param[in] height          Height of the image
- * @param[in] x               X position of the wanted pixel
- * @param[in] y               Y position of the wanted pixel
- *
- * @return The pixel at (x, y) using bilinear interpolation.
- */
-template <typename T>
-inline uint8_t pixel_bilinear_c1_clamp(const T *first_pixel_ptr, size_t stride, size_t width, size_t height, float x, float y)
-{
-    ARM_COMPUTE_ERROR_ON(first_pixel_ptr == nullptr);
-
-    x = std::max(-1.f, std::min(x, static_cast<float>(width)));
-    y = std::max(-1.f, std::min(y, static_cast<float>(height)));
-
-    const float xi = std::floor(x);
-    const float yi = std::floor(y);
-
-    const float dx = x - xi;
-    const float dy = y - yi;
-
-    if(dx == 0.0f)
-    {
-        if(dy == 0.0f)
-        {
-            return static_cast<T>(first_pixel_ptr[static_cast<int32_t>(xi) + static_cast<int32_t>(yi) * stride]);
-        }
-        return delta_linear_c1_y(first_pixel_ptr + static_cast<int32_t>(xi) + static_cast<int32_t>(yi) * stride, stride, dy);
-    }
-    if(dy == 0.0f)
-    {
-        return delta_linear_c1_x(first_pixel_ptr + static_cast<int32_t>(xi) + static_cast<int32_t>(yi) * stride, dx);
-    }
-    return delta_bilinear_c1(first_pixel_ptr + static_cast<int32_t>(xi) + static_cast<int32_t>(yi) * stride, stride, dx, dy);
-}
-
-/** Return the pixel at (x,y) using area interpolation by clamping when out of borders. The image must be single channel U8
- *
- * @note The interpolation area depends on the width and height ration of the input and output images
- * @note Currently average of the contributing pixels is calculated
- *
- * @param[in] first_pixel_ptr Pointer to the first pixel of a single channel U8 image.
- * @param[in] stride          Stride in bytes of the image
- * @param[in] width           Width of the image
- * @param[in] height          Height of the image
- * @param[in] wr              Width ratio among the input image width and output image width.
- * @param[in] hr              Height ratio among the input image height and output image height.
- * @param[in] x               X position of the wanted pixel
- * @param[in] y               Y position of the wanted pixel
- *
- * @return The pixel at (x, y) using area interpolation.
- */
-inline uint8_t pixel_area_c1u8_clamp(const uint8_t *first_pixel_ptr, size_t stride, size_t width, size_t height, float wr, float hr, int x, int y);
-
 /** Iterator updated by @ref execute_window_loop for each window element */
 class Iterator
 {
@@ -421,179 +114,6 @@
 template <typename L, typename... Ts>
 inline void execute_window_loop(const Window &w, L &&lambda_function, Ts &&... iterators);
 
-/** Update window and padding size for each of the access patterns.
- *
- * First the window size is reduced based on all access patterns that are not
- * allowed to modify the padding of the underlying tensor. Then the padding of
- * the remaining tensors is increased to match the window.
- *
- * @param[in] win      Window that is used by the kernel.
- * @param[in] patterns Access patterns used to calculate the final window and padding.
- *
- * @return True if the window has been changed. Changes to the padding do not
- *         influence the returned value.
- */
-template <typename... Ts>
-bool update_window_and_padding(Window &win, Ts &&... patterns)
-{
-    bool window_changed = false;
-
-    utility::for_each([&](const IAccessWindow & w)
-    {
-        window_changed |= w.update_window_if_needed(win);
-    },
-    patterns...);
-
-    bool padding_changed = false;
-
-    utility::for_each([&](IAccessWindow & w)
-    {
-        padding_changed |= w.update_padding_if_needed(win);
-    },
-    patterns...);
-
-    return window_changed;
-}
-
-/** Calculate the maximum window for a given tensor shape and border setting
- *
- * @param[in] valid_region Valid region object defining the shape of the tensor space for which the window is created.
- * @param[in] steps        (Optional) Number of elements processed for each step.
- * @param[in] skip_border  (Optional) If true exclude the border region from the window.
- * @param[in] border_size  (Optional) Border size.
- *
- * @return The maximum window the kernel can be executed on.
- */
-Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps = Steps(), bool skip_border = false, BorderSize border_size = BorderSize());
-
-/** Calculate the maximum window for a given tensor shape and border setting
- *
- * @param[in] info        Tensor info object defining the shape of the object for which the window is created.
- * @param[in] steps       (Optional) Number of elements processed for each step.
- * @param[in] skip_border (Optional) If true exclude the border region from the window.
- * @param[in] border_size (Optional) Border size.
- *
- * @return The maximum window the kernel can be executed on.
- */
-inline Window calculate_max_window(const ITensorInfo &info, const Steps &steps = Steps(), bool skip_border = false, BorderSize border_size = BorderSize())
-{
-    return calculate_max_window(info.valid_region(), steps, skip_border, border_size);
-}
-
-/** Calculate the maximum window used by a horizontal kernel for a given tensor shape and border setting
- *
- * @param[in] valid_region Valid region object defining the shape of the tensor space for which the window is created.
- * @param[in] steps        (Optional) Number of elements processed for each step.
- * @param[in] skip_border  (Optional) If true exclude the border region from the window.
- * @param[in] border_size  (Optional) Border size. The border region will be excluded from the window.
- *
- * @return The maximum window the kernel can be executed on.
- */
-Window calculate_max_window_horizontal(const ValidRegion &valid_region, const Steps &steps = Steps(), bool skip_border = false, BorderSize border_size = BorderSize());
-
-/** Calculate the maximum window used by a horizontal kernel for a given tensor shape and border setting
- *
- * @param[in] info        Tensor info object defining the shape of the object for which the window is created.
- * @param[in] steps       (Optional) Number of elements processed for each step.
- * @param[in] skip_border (Optional) If true exclude the border region from the window.
- * @param[in] border_size (Optional) Border size.
- *
- * @return The maximum window the kernel can be executed on.
- */
-inline Window calculate_max_window_horizontal(const ITensorInfo &info, const Steps &steps = Steps(), bool skip_border = false, BorderSize border_size = BorderSize())
-{
-    return calculate_max_window_horizontal(info.valid_region(), steps, skip_border, border_size);
-}
-
-/** Calculate the maximum window for a given tensor shape and border setting. The window will also includes the border.
- *
- * @param[in] valid_region Valid region object defining the shape of the tensor space for which the window is created.
- * @param[in] steps        (Optional) Number of elements processed for each step.
- * @param[in] border_size  (Optional) Border size. The border region will be included in the window.
- *
- * @return The maximum window the kernel can be executed on.
- */
-Window calculate_max_enlarged_window(const ValidRegion &valid_region, const Steps &steps = Steps(), BorderSize border_size = BorderSize());
-
-/** Calculate the maximum window for a given tensor shape and border setting. The window will also includes the border.
- *
- * @param[in] info        Tensor info object defining the shape of the object for which the window is created.
- * @param[in] steps       (Optional) Number of elements processed for each step.
- * @param[in] border_size (Optional) Border size. The border region will be included in the window.
- *
- * @return The maximum window the kernel can be executed on.
- */
-inline Window calculate_max_enlarged_window(const ITensorInfo &info, const Steps &steps = Steps(), BorderSize border_size = BorderSize())
-{
-    return calculate_max_enlarged_window(info.valid_region(), steps, border_size);
-}
-
-/** Intersect multiple valid regions.
- *
- * @param[in] regions Valid regions.
- *
- * @return Intersection of all regions.
- */
-template <typename... Ts>
-ValidRegion intersect_valid_regions(const Ts &... regions)
-{
-    auto intersect = [](const ValidRegion & r1, const ValidRegion & r2) -> ValidRegion
-    {
-        ValidRegion region;
-
-        for(size_t d = 0; d < std::min(r1.anchor.num_dimensions(), r2.anchor.num_dimensions()); ++d)
-        {
-            region.anchor.set(d, std::max(r1.anchor[d], r2.anchor[d]));
-        }
-
-        for(size_t d = 0; d < std::min(r1.shape.num_dimensions(), r2.shape.num_dimensions()); ++d)
-        {
-            region.shape.set(d, std::min(r1.shape[d], r2.shape[d]));
-        }
-
-        return region;
-    };
-
-    return utility::foldl(intersect, regions...);
-}
-
-/** Create a strides object based on the provided strides and the tensor dimensions.
- *
- * @param[in] info          Tensor info object providing the shape of the tensor for unspecified strides.
- * @param[in] stride_x      Stride to be used in X dimension (in bytes).
- * @param[in] fixed_strides Strides to be used in higher dimensions starting at Y (in bytes).
- *
- * @return Strides object based on the specified strides. Missing strides are
- *         calculated based on the tensor shape and the strides of lower dimensions.
- */
-template <typename T, typename... Ts>
-inline Strides compute_strides(const ITensorInfo &info, T stride_x, Ts &&... fixed_strides)
-{
-    const TensorShape &shape = info.tensor_shape();
-
-    // Create strides object
-    Strides strides(stride_x, fixed_strides...);
-
-    for(size_t i = 1 + sizeof...(Ts); i < info.num_dimensions(); ++i)
-    {
-        strides.set(i, shape[i - 1] * strides[i - 1]);
-    }
-
-    return strides;
-}
-
-/** Create a strides object based on the tensor dimensions.
- *
- * @param[in] info Tensor info object used to compute the strides.
- *
- * @return Strides object based on element size and tensor shape.
- */
-template <typename... Ts>
-inline Strides compute_strides(const ITensorInfo &info)
-{
-    return compute_strides(info, info.element_size());
-}
-
 /** Permutes given Dimensions according to a permutation vector
  *
  * @warning Validity of permutation is not checked
@@ -629,79 +149,6 @@
     }
 }
 
-/** Auto initialize the tensor info (shape, number of channels and data type) if the current assignment is empty.
- *
- * @param[in,out] info              Tensor info used to check and assign.
- * @param[in]     shape             New shape.
- * @param[in]     num_channels      New number of channels.
- * @param[in]     data_type         New data type
- * @param[in]     quantization_info (Optional) New quantization info
- *
- * @return True if the tensor info has been initialized
- */
-bool auto_init_if_empty(ITensorInfo       &info,
-                        const TensorShape &shape,
-                        int num_channels, DataType data_type,
-                        QuantizationInfo quantization_info = QuantizationInfo());
-
-/** Auto initialize the tensor info using another tensor info.
- *
- * @param info_sink   Tensor info used to check and assign
- * @param info_source Tensor info used to assign
- *
- * @return True if the tensor info has been initialized
- */
-bool auto_init_if_empty(ITensorInfo &info_sink, const ITensorInfo &info_source);
-
-/** Set the shape to the specified value if the current assignment is empty.
- *
- * @param[in,out] info  Tensor info used to check and assign.
- * @param[in]     shape New shape.
- *
- * @return True if the shape has been changed.
- */
-bool set_shape_if_empty(ITensorInfo &info, const TensorShape &shape);
-
-/** Set the format, data type and number of channels to the specified value if
- * the current data type is unknown.
- *
- * @param[in,out] info   Tensor info used to check and assign.
- * @param[in]     format New format.
- *
- * @return True if the format has been changed.
- */
-bool set_format_if_unknown(ITensorInfo &info, Format format);
-
-/** Set the data type and number of channels to the specified value if
- * the current data type is unknown.
- *
- * @param[in,out] info      Tensor info used to check and assign.
- * @param[in]     data_type New data type.
- *
- * @return True if the data type has been changed.
- */
-bool set_data_type_if_unknown(ITensorInfo &info, DataType data_type);
-
-/** Set the data layout to the specified value if
- * the current data layout is unknown.
- *
- * @param[in,out] info        Tensor info used to check and assign.
- * @param[in]     data_layout New data layout.
- *
- * @return True if the data type has been changed.
- */
-bool set_data_layout_if_unknown(ITensorInfo &info, DataLayout data_layout);
-
-/** Set the quantization info to the specified value if
- * the current quantization info is empty and the data type of asymmetric quantized type
- *
- * @param[in,out] info              Tensor info used to check and assign.
- * @param[in]     quantization_info Quantization info
- *
- * @return True if the quantization info has been changed.
- */
-bool set_quantization_info_if_empty(ITensorInfo &info, QuantizationInfo quantization_info);
-
 /** Helper function to calculate the Valid Region for Scale.
  *
  * @param[in] src_info           Input tensor info used to check.
@@ -751,21 +198,6 @@
  */
 inline DataLayoutDimension get_index_data_layout_dimension(const DataLayout data_layout, const size_t index);
 
-/** Calculate the normalization dimension index for a given normalization type
- *
- * @param[in] layout Data layout of the input and output tensor
- * @param[in] info   Normalization info
- *
- * @return Normalization dimension index
- */
-inline unsigned int get_normalization_dimension_index(DataLayout layout, const NormalizationLayerInfo &info)
-{
-    const unsigned int width_idx   = get_data_layout_dimension_index(layout, DataLayoutDimension::WIDTH);
-    const unsigned int channel_idx = get_data_layout_dimension_index(layout, DataLayoutDimension::CHANNEL);
-
-    return info.is_in_map() ? width_idx : channel_idx;
-}
-
 /** Calculate the number of output tiles required by Winograd Convolution layer. This utility function can be used by the Winograd input transform
  *  to know the number of tiles on the x and y direction
  *
@@ -814,49 +246,6 @@
     }
     return coords;
 }
-
-/** Given an integer value, this function returns the next power of two
- *
- * @param[in] x Input value
- *
- * @return the next power of two
- */
-inline unsigned int get_next_power_two(unsigned int x)
-{
-    // Decrement by 1
-    x--;
-
-    // Shift right by 1
-    x |= x >> 1u;
-    // Shift right by 2
-    x |= x >> 2u;
-    // Shift right by 4
-    x |= x >> 4u;
-    // Shift right by 8
-    x |= x >> 8u;
-    // Shift right by 16
-    x |= x >> 16u;
-
-    // Increment by 1
-    x++;
-
-    return x;
-}
-
-/** Given a softmax axis, this function returns the permutation vector required to put the axis to the front
- *
- * @note This function assumes a tensor rank <= 4
- *
- * Axis selects the dimension on which softmax is performed.
- * E.g. For input of shape 4x5x6 and axis=1, softmax will be applied to 4x6=24 vectors of size 5.
- * Interally softmax kernels is always performed on the first dimension (front dimension), therefore permutation is
- * required to put the dimension specified by @p axis to the first dimension.
- *
- * @param[in] axis Axis on which to perform softmax. Supported: 1, 2, 3 (0 implies no permutation needed)
- *
- * @return the permutation vector
- */
-PermutationVector get_permutation_vector_from_softmax_axis(size_t axis);
 } // namespace arm_compute
 
 #include "arm_compute/core/Helpers.inl"
diff --git a/arm_compute/core/Helpers.inl b/arm_compute/core/Helpers.inl
index 5613e8c..a960876 100644
--- a/arm_compute/core/Helpers.inl
+++ b/arm_compute/core/Helpers.inl
@@ -22,58 +22,12 @@
  * SOFTWARE.
  */
 #include "arm_compute/core/Error.h"
-#include "arm_compute/core/Validate.h"
 
 #include <cmath>
 #include <numeric>
 
 namespace arm_compute
 {
-inline uint8_t pixel_area_c1u8_clamp(const uint8_t *first_pixel_ptr, size_t stride, size_t width, size_t height, float wr, float hr, int x, int y)
-{
-    ARM_COMPUTE_ERROR_ON(first_pixel_ptr == nullptr);
-
-    // Calculate sampling position
-    float in_x = (x + 0.5f) * wr - 0.5f;
-    float in_y = (y + 0.5f) * hr - 0.5f;
-
-    // Get bounding box offsets
-    int x_from = std::floor(x * wr - 0.5f - in_x);
-    int y_from = std::floor(y * hr - 0.5f - in_y);
-    int x_to   = std::ceil((x + 1) * wr - 0.5f - in_x);
-    int y_to   = std::ceil((y + 1) * hr - 0.5f - in_y);
-
-    // Clamp position to borders
-    in_x = std::max(-1.f, std::min(in_x, static_cast<float>(width)));
-    in_y = std::max(-1.f, std::min(in_y, static_cast<float>(height)));
-
-    // Clamp bounding box offsets to borders
-    x_from = ((in_x + x_from) < -1) ? -1 : x_from;
-    y_from = ((in_y + y_from) < -1) ? -1 : y_from;
-    x_to   = ((in_x + x_to) > width) ? (width - in_x) : x_to;
-    y_to   = ((in_y + y_to) > height) ? (height - in_y) : y_to;
-
-    // Get pixel index
-    const int xi = std::floor(in_x);
-    const int yi = std::floor(in_y);
-
-    // Bounding box elements in each dimension
-    const int x_elements = (x_to - x_from + 1);
-    const int y_elements = (y_to - y_from + 1);
-    ARM_COMPUTE_ERROR_ON(x_elements == 0 || y_elements == 0);
-
-    // Sum pixels in area
-    int sum = 0;
-    for(int j = yi + y_from, je = yi + y_to; j <= je; ++j)
-    {
-        const uint8_t *ptr = first_pixel_ptr + j * stride + xi + x_from;
-        sum                = std::accumulate(ptr, ptr + x_elements, sum);
-    }
-
-    // Return average
-    return sum / (x_elements * y_elements);
-}
-
 template <size_t dimension>
 struct IncrementIterators
 {
@@ -199,94 +153,6 @@
     }
 }
 
-inline bool auto_init_if_empty(ITensorInfo       &info,
-                               const TensorShape &shape,
-                               int                num_channels,
-                               DataType           data_type,
-                               QuantizationInfo   quantization_info)
-{
-    if(info.tensor_shape().total_size() == 0)
-    {
-        info.set_data_type(data_type);
-        info.set_num_channels(num_channels);
-        info.set_tensor_shape(shape);
-        info.set_quantization_info(quantization_info);
-        return true;
-    }
-
-    return false;
-}
-
-inline bool auto_init_if_empty(ITensorInfo &info_sink, const ITensorInfo &info_source)
-{
-    if(info_sink.tensor_shape().total_size() == 0)
-    {
-        info_sink.set_data_type(info_source.data_type());
-        info_sink.set_num_channels(info_source.num_channels());
-        info_sink.set_tensor_shape(info_source.tensor_shape());
-        info_sink.set_quantization_info(info_source.quantization_info());
-        info_sink.set_data_layout(info_source.data_layout());
-        return true;
-    }
-
-    return false;
-}
-
-inline bool set_shape_if_empty(ITensorInfo &info, const TensorShape &shape)
-{
-    if(info.tensor_shape().total_size() == 0)
-    {
-        info.set_tensor_shape(shape);
-        return true;
-    }
-
-    return false;
-}
-
-inline bool set_format_if_unknown(ITensorInfo &info, Format format)
-{
-    if(info.data_type() == DataType::UNKNOWN)
-    {
-        info.set_format(format);
-        return true;
-    }
-
-    return false;
-}
-
-inline bool set_data_type_if_unknown(ITensorInfo &info, DataType data_type)
-{
-    if(info.data_type() == DataType::UNKNOWN)
-    {
-        info.set_data_type(data_type);
-        return true;
-    }
-
-    return false;
-}
-
-inline bool set_data_layout_if_unknown(ITensorInfo &info, DataLayout data_layout)
-{
-    if(info.data_layout() == DataLayout::UNKNOWN)
-    {
-        info.set_data_layout(data_layout);
-        return true;
-    }
-
-    return false;
-}
-
-inline bool set_quantization_info_if_empty(ITensorInfo &info, QuantizationInfo quantization_info)
-{
-    if(info.quantization_info().empty() && (is_data_type_quantized_asymmetric(info.data_type())))
-    {
-        info.set_quantization_info(quantization_info);
-        return true;
-    }
-
-    return false;
-}
-
 inline Coordinates index2coords(const TensorShape &shape, int index)
 {
     int num_elements = shape.total_size();
diff --git a/arm_compute/core/ITensorInfo.h b/arm_compute/core/ITensorInfo.h
index c5f0949..3eb7239 100644
--- a/arm_compute/core/ITensorInfo.h
+++ b/arm_compute/core/ITensorInfo.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,8 +29,8 @@
 #include "arm_compute/core/TensorShape.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/utils/misc/ICloneable.h"
 #include "arm_compute/core/utils/misc/Utility.h"
+#include "support/ICloneable.h"
 
 #include <cstddef>
 
diff --git a/arm_compute/core/NEON/kernels/NEDepthwiseConvolutionLayerNativeKernel.h b/arm_compute/core/NEON/kernels/NEDepthwiseConvolutionLayerNativeKernel.h
index 335a70f..eba1737 100644
--- a/arm_compute/core/NEON/kernels/NEDepthwiseConvolutionLayerNativeKernel.h
+++ b/arm_compute/core/NEON/kernels/NEDepthwiseConvolutionLayerNativeKernel.h
@@ -26,6 +26,7 @@
 
 #include "arm_compute/core/NEON/INEKernel.h"
 #include "arm_compute/core/utils/misc/Traits.h"
+#include "support/Requires.h"
 
 #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC
 #include <arm_neon.h>
diff --git a/arm_compute/core/SubTensorInfo.h b/arm_compute/core/SubTensorInfo.h
index f604f55..6654ccf 100644
--- a/arm_compute/core/SubTensorInfo.h
+++ b/arm_compute/core/SubTensorInfo.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -31,7 +31,6 @@
 #include "arm_compute/core/Strides.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/TensorShape.h"
-#include "arm_compute/core/Validate.h"
 
 #include <cstddef>
 #include <memory>
diff --git a/arm_compute/core/utils/math/SafeOps.h b/arm_compute/core/utils/math/SafeOps.h
index 4f81cf4..c222c65 100644
--- a/arm_compute/core/utils/math/SafeOps.h
+++ b/arm_compute/core/utils/math/SafeOps.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -25,7 +25,9 @@
 #define ARM_COMPUTE_UTILS_MATH_SAFE_OPS
 
 #include "arm_compute/core/Error.h"
-#include "arm_compute/core/utils/misc/Requires.h"
+#include "support/Requires.h"
+
+#include <limits>
 
 namespace arm_compute
 {
diff --git a/arm_compute/graph/TensorDescriptor.h b/arm_compute/graph/TensorDescriptor.h
index 6c6f99d..de67289 100644
--- a/arm_compute/graph/TensorDescriptor.h
+++ b/arm_compute/graph/TensorDescriptor.h
@@ -26,7 +26,7 @@
 
 #include "arm_compute/graph/Types.h"
 
-#include "arm_compute/core/utils/misc/ICloneable.h"
+#include "support/ICloneable.h"
 #include "support/MemorySupport.h"
 
 #include <memory>
diff --git a/arm_compute/graph/backends/FunctionHelpers.h b/arm_compute/graph/backends/FunctionHelpers.h
index e8395e4..e2904af 100644
--- a/arm_compute/graph/backends/FunctionHelpers.h
+++ b/arm_compute/graph/backends/FunctionHelpers.h
@@ -37,7 +37,7 @@
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensorInfo.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "support/Cast.h"
 
 namespace arm_compute
 {
diff --git a/arm_compute/runtime/CL/tuners/CLLWSList.h b/arm_compute/runtime/CL/tuners/CLLWSList.h
index 7ce10ac..48f3f3f 100644
--- a/arm_compute/runtime/CL/tuners/CLLWSList.h
+++ b/arm_compute/runtime/CL/tuners/CLLWSList.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,7 +29,8 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/runtime/CL/CLTunerTypes.h"
 #include "support/ToolchainSupport.h"
-#include <memory>
+
+#include "support/MemorySupport.h"
 
 namespace arm_compute
 {
diff --git a/arm_compute/runtime/CL/tuners/Tuners.h b/arm_compute/runtime/CL/tuners/Tuners.h
index 274f13d..dd1c62a 100644
--- a/arm_compute/runtime/CL/tuners/Tuners.h
+++ b/arm_compute/runtime/CL/tuners/Tuners.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -27,6 +27,8 @@
 #include "arm_compute/runtime/CL/tuners/BifrostTuner.h"
 #include "arm_compute/runtime/CL/tuners/MidgardTuner.h"
 
+#include "support/MemorySupport.h"
+
 #include <memory>
 
 namespace arm_compute
diff --git a/arm_compute/runtime/CPP/functions/CPPSplit.h b/arm_compute/runtime/CPP/functions/CPPSplit.h
index 7929f14..b2b4d07 100644
--- a/arm_compute/runtime/CPP/functions/CPPSplit.h
+++ b/arm_compute/runtime/CPP/functions/CPPSplit.h
@@ -106,7 +106,10 @@
 
             // Output auto inizialitation if not yet initialized
             TensorInfo tmp_output_info = *output->clone();
-            auto_init_if_empty(tmp_output_info, input->clone()->set_is_resizable(true).set_tensor_shape(output_shape));
+            if(tmp_output_info.tensor_shape().total_size() == 0)
+            {
+                tmp_output_info = input->clone()->set_is_resizable(true).set_tensor_shape(output_shape);
+            }
 
             // Update coordinate on axis
             start_coords.set(axis, axis_offset);
diff --git a/arm_compute/runtime/NEON/NEFunctions.h b/arm_compute/runtime/NEON/NEFunctions.h
index 763294e..a97fa3b 100644
--- a/arm_compute/runtime/NEON/NEFunctions.h
+++ b/arm_compute/runtime/NEON/NEFunctions.h
@@ -136,7 +136,6 @@
 #include "arm_compute/runtime/NEON/functions/NEScale.h"
 #include "arm_compute/runtime/NEON/functions/NEScharr3x3.h"
 #include "arm_compute/runtime/NEON/functions/NESelect.h"
-#include "arm_compute/runtime/NEON/functions/NESimpleAssemblyFunction.h"
 #include "arm_compute/runtime/NEON/functions/NESlice.h"
 #include "arm_compute/runtime/NEON/functions/NESobel3x3.h"
 #include "arm_compute/runtime/NEON/functions/NESobel5x5.h"
diff --git a/arm_compute/runtime/NEON/functions/NEConcatenateLayer.h b/arm_compute/runtime/NEON/functions/NEConcatenateLayer.h
index 1d703ae..82b4517 100644
--- a/arm_compute/runtime/NEON/functions/NEConcatenateLayer.h
+++ b/arm_compute/runtime/NEON/functions/NEConcatenateLayer.h
@@ -28,8 +28,8 @@
 
 #include "arm_compute/core/NEON/INEKernel.h"
 #include "arm_compute/core/Types.h"
-#include "arm_compute/core/utils/misc/Requires.h"
 #include "arm_compute/runtime/NEON/INEOperator.h"
+#include "support/Requires.h"
 
 #include <memory>
 #include <vector>
diff --git a/docs/ComputeLibrary.dir b/docs/ComputeLibrary.dir
index 270e7fa..52f1a80 100644
--- a/docs/ComputeLibrary.dir
+++ b/docs/ComputeLibrary.dir
@@ -31,10 +31,6 @@
  *  @brief Wrapper to configure the Khronos OpenCL C++ header
  */
 
-/** @dir arm_compute/core/CL/gemm
- *  @brief Folder containing all the configuration files for GEMM
- */
-
 /** @dir arm_compute/core/CL/kernels
  * @brief Folder containing all the OpenCL kernels
  */
@@ -283,6 +279,10 @@
  *  @brief Scalar operations
  */
 
+/** @dir src/core/CL/gemm
+ *  @brief Folder containing all the configuration files for GEMM
+ */
+
 /** @dir src/core/CL/cl_kernels
  *  @brief All the OpenCL kernels
  */
diff --git a/examples/gemm_tuner/GemmTunerHelpers.h b/examples/gemm_tuner/GemmTunerHelpers.h
new file mode 100644
index 0000000..23cf14c
--- /dev/null
+++ b/examples/gemm_tuner/GemmTunerHelpers.h
@@ -0,0 +1,51 @@
+/*
+ * Copyright (c) 2020 Arm Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef EXAMPLES_GEMMTUNERHELPERS_H
+#define EXAMPLES_GEMMTUNERHELPERS_H
+
+#include "arm_compute/core/CL/CLHelpers.h"
+#include "arm_compute/core/CL/CLKernelLibrary.h"
+
+namespace examples
+{
+namespace gemm_tuner_helpers
+{
+void update_padding_for_cl_image(arm_compute::ITensorInfo *tensor)
+{
+    constexpr unsigned int num_floats_per_pixel = 4;
+
+    const unsigned int stride_y_in_elements = tensor->strides_in_bytes()[1] / tensor->element_size();
+    const unsigned int pixel_aligment       = arm_compute::get_cl_image_pitch_alignment(
+                                                  arm_compute::CLKernelLibrary::get().get_device());
+    const unsigned int row_pitch_alignment = pixel_aligment * num_floats_per_pixel;
+    const unsigned int round_up_width =
+        ((stride_y_in_elements + row_pitch_alignment - 1) / row_pitch_alignment) * row_pitch_alignment;
+    const unsigned int padding = round_up_width - stride_y_in_elements;
+
+    tensor->extend_padding(arm_compute::PaddingSize(0, padding, 0, 0));
+}
+} // namespace gemm_tuner_helpers
+} // namespace examples
+
+#endif /* EXAMPLES_GEMMTUNERHELPERS_H */
diff --git a/examples/gemm_tuner/cl_gemm_reshaped.cpp b/examples/gemm_tuner/cl_gemm_reshaped.cpp
index 9c6568c..92fa990 100644
--- a/examples/gemm_tuner/cl_gemm_reshaped.cpp
+++ b/examples/gemm_tuner/cl_gemm_reshaped.cpp
@@ -25,8 +25,6 @@
 #error "This example needs to be built with -DARM_COMPUTE_CL"
 #endif /* ARM_COMPUTE_CL */
 
-#include "CommonGemmExampleOptions.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
 #include "arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyReshapedKernel.h"
 #include "arm_compute/core/CL/kernels/CLGEMMReshapeLHSMatrixKernel.h"
 #include "arm_compute/core/Helpers.h"
@@ -36,6 +34,8 @@
 #include "arm_compute/runtime/CL/CLFunctions.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "arm_compute/runtime/CL/CLTuner.h"
+#include "examples/gemm_tuner/CommonGemmExampleOptions.h"
+#include "examples/gemm_tuner/GemmTunerHelpers.h"
 #include "tests/CL/Helper.h"
 #include "utils/Utils.h"
 #include "utils/command_line/CommandLineOptions.h"
@@ -254,14 +254,14 @@
         kernel_info.activation_info         = act_info;
 
         // Initialise lhs_reshaped tensor info
-        auto_init_if_empty(*lhs_reshaped.info(), lhs.info()->clone()->set_tensor_shape(compute_lhs_reshaped_shape(*lhs.info(), lhs_info)));
+        lhs_reshaped.allocator()->init(TensorInfo(compute_lhs_reshaped_shape(*lhs.info(), lhs_info), 1, params.data_type));
 
         // Initialise rhs_reshaped tensor info
-        auto_init_if_empty(*rhs_reshaped.info(), rhs.info()->clone()->set_tensor_shape(compute_rhs_reshaped_shape(*rhs.info(), rhs_info)));
+        rhs_reshaped.allocator()->init(TensorInfo(compute_rhs_reshaped_shape(*rhs.info(), rhs_info), 1, params.data_type));
 
         if(rhs_info.export_to_cl_image)
         {
-            arm_compute::cl_gemm::update_padding_for_cl_image(rhs_reshaped.info());
+            examples::gemm_tuner_helpers::update_padding_for_cl_image(rhs_reshaped.info());
         }
 
         // Validate argments
diff --git a/examples/gemm_tuner/cl_gemm_reshaped_rhs_only.cpp b/examples/gemm_tuner/cl_gemm_reshaped_rhs_only.cpp
index f814c54..3a76001 100644
--- a/examples/gemm_tuner/cl_gemm_reshaped_rhs_only.cpp
+++ b/examples/gemm_tuner/cl_gemm_reshaped_rhs_only.cpp
@@ -26,7 +26,7 @@
 #endif /* ARM_COMPUTE_CL */
 
 #include "CommonGemmExampleOptions.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
+#include "GemmTunerHelpers.h"
 #include "arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/KernelDescriptors.h"
@@ -224,11 +224,11 @@
         kernel_info.activation_info         = act_info;
 
         // Initialise rhs_reshaped tensor info
-        auto_init_if_empty(*rhs_reshaped.info(), rhs.info()->clone()->set_tensor_shape(compute_rhs_reshaped_shape(*rhs.info(), rhs_info)));
+        rhs_reshaped.allocator()->init(TensorInfo(compute_rhs_reshaped_shape(*rhs.info(), rhs_info), 1, params.data_type));
 
         if(rhs_info.export_to_cl_image)
         {
-            arm_compute::cl_gemm::update_padding_for_cl_image(rhs_reshaped.info());
+            examples::gemm_tuner_helpers::update_padding_for_cl_image(rhs_reshaped.info());
         }
 
         // Validate argments
diff --git a/scripts/clang_tidy_rules.py b/scripts/clang_tidy_rules.py
index 5e13aa0..ce467f8 100755
--- a/scripts/clang_tidy_rules.py
+++ b/scripts/clang_tidy_rules.py
@@ -111,11 +111,15 @@
                ("NEWinogradLayerKernel.cpp" in line and "use '= default' to define a trivial destructor" in line) or
                ("NEGEMMLowpMatrixMultiplyCore.cpp" in line and "constructor does not initialize these fields" in line) or
                ("NEGEMMLowpAssemblyMatrixMultiplyCore" in line and "constructor does not initialize these fields" in line) or
+               ("NEDepthwiseConvolutionLayerNativeKernel" in line and re.search(r"parameter '[^']+' is unused", line)) or
+               ("NEDepthwiseConvolutionAssemblyDispatch" in line and re.search(r"parameter '[^']+' is unused", line)) or
                ("CPUUtils.cpp" in line and "consider replacing 'unsigned long' with 'uint64'" in line) or
                ("CPUUtils.cpp" in line and "parameter 'cpusv' is unused" in line) or
                ("CPUUtils.cpp" in line and "warning: uninitialized record type" in line) or
                ("GCKernelLibrary.cpp" in line and "warning: do not declare C-style arrays" in line) or
                ("Utils.h" in line and "warning: Use of zero-allocated memory" in line) or
+               ("NEDepthwiseConvolutionLayerNativeKernel.cpp" in line and "misc-non-private-member-variables-in-classes" in line) or # This is to prevent false positive, should be reassessed with the newer clang-tidy
+               ("NEDepthwiseConvolutionLayerNativeKernel.cpp" in line and "cppcoreguidelines-pro-type-member-init" in line) or # This is to prevent false positive, should be reassessed with the newer clang-tidy
                "3rdparty" in line):
                 print_context=False
                 continue
diff --git a/src/core/AccessWindowAutoPadding.cpp b/src/core/AccessWindowAutoPadding.cpp
index 85c5b27..ca2f7d2 100644
--- a/src/core/AccessWindowAutoPadding.cpp
+++ b/src/core/AccessWindowAutoPadding.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/AccessWindowAutoPadding.h"
+#include "src/core/AccessWindowAutoPadding.h"
 
 #include "arm_compute/core/ITensorInfo.h"
 #include "arm_compute/core/Window.h"
diff --git a/arm_compute/core/AccessWindowAutoPadding.h b/src/core/AccessWindowAutoPadding.h
similarity index 98%
rename from arm_compute/core/AccessWindowAutoPadding.h
rename to src/core/AccessWindowAutoPadding.h
index 12d6553..b8d1508 100644
--- a/arm_compute/core/AccessWindowAutoPadding.h
+++ b/src/core/AccessWindowAutoPadding.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/src/core/AccessWindowStatic.cpp b/src/core/AccessWindowStatic.cpp
index 10e88b8..0607011 100644
--- a/src/core/AccessWindowStatic.cpp
+++ b/src/core/AccessWindowStatic.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2018 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/AccessWindowStatic.h"
+#include "src/core/AccessWindowStatic.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensorInfo.h"
diff --git a/arm_compute/core/AccessWindowStatic.h b/src/core/AccessWindowStatic.h
similarity index 98%
rename from arm_compute/core/AccessWindowStatic.h
rename to src/core/AccessWindowStatic.h
index 1f2ca1b..f7d43cb 100644
--- a/arm_compute/core/AccessWindowStatic.h
+++ b/src/core/AccessWindowStatic.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/src/core/AccessWindowTranspose.cpp b/src/core/AccessWindowTranspose.cpp
index 4c03ca1..d8bd4c4 100644
--- a/src/core/AccessWindowTranspose.cpp
+++ b/src/core/AccessWindowTranspose.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2018 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/AccessWindowTranspose.h"
+#include "src/core/AccessWindowTranspose.h"
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensorInfo.h"
diff --git a/arm_compute/core/AccessWindowTranspose.h b/src/core/AccessWindowTranspose.h
similarity index 97%
rename from arm_compute/core/AccessWindowTranspose.h
rename to src/core/AccessWindowTranspose.h
index 8570909..0306076 100644
--- a/arm_compute/core/AccessWindowTranspose.h
+++ b/src/core/AccessWindowTranspose.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/arm_compute/core/CL/CLValidate.h b/src/core/CL/CLValidate.h
similarity index 98%
rename from arm_compute/core/CL/CLValidate.h
rename to src/core/CL/CLValidate.h
index 3f8b76b..cbbdf2d 100644
--- a/arm_compute/core/CL/CLValidate.h
+++ b/src/core/CL/CLValidate.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/arm_compute/core/CL/ICLGEMMKernelConfiguration.h b/src/core/CL/ICLGEMMKernelConfiguration.h
similarity index 98%
rename from arm_compute/core/CL/ICLGEMMKernelConfiguration.h
rename to src/core/CL/ICLGEMMKernelConfiguration.h
index 90600ef..ac0e7ab 100644
--- a/arm_compute/core/CL/ICLGEMMKernelConfiguration.h
+++ b/src/core/CL/ICLGEMMKernelConfiguration.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/src/core/CL/ICLKernel.cpp b/src/core/CL/ICLKernel.cpp
index be63374..f91510b 100644
--- a/src/core/CL/ICLKernel.cpp
+++ b/src/core/CL/ICLKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,14 +23,9 @@
  */
 #include "arm_compute/core/CL/ICLKernel.h"
 
-#include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/Utils.h"
 
 #include <cstddef>
 
diff --git a/src/core/CL/ICLSimple2DKernel.cpp b/src/core/CL/ICLSimple2DKernel.cpp
index ce95495..dfef582 100644
--- a/src/core/CL/ICLSimple2DKernel.cpp
+++ b/src/core/CL/ICLSimple2DKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2018 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,10 +23,7 @@
  */
 #include "arm_compute/core/CL/ICLSimple2DKernel.h"
 
-#include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute;
 
diff --git a/src/core/CL/ICLSimpleKernel.cpp b/src/core/CL/ICLSimpleKernel.cpp
index d2f09a3..90b5be8 100644
--- a/src/core/CL/ICLSimpleKernel.cpp
+++ b/src/core/CL/ICLSimpleKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2018 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -27,6 +27,7 @@
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute;
 
diff --git a/src/core/CL/gemm/CLGEMMHelpers.cpp b/src/core/CL/gemm/CLGEMMHelpers.cpp
index 0a4a4ad..877bf1e 100644
--- a/src/core/CL/gemm/CLGEMMHelpers.cpp
+++ b/src/core/CL/gemm/CLGEMMHelpers.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
diff --git a/arm_compute/core/CL/gemm/CLGEMMHelpers.h b/src/core/CL/gemm/CLGEMMHelpers.h
similarity index 100%
rename from arm_compute/core/CL/gemm/CLGEMMHelpers.h
rename to src/core/CL/gemm/CLGEMMHelpers.h
diff --git a/arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfiguration.h b/src/core/CL/gemm/native/CLGEMMNativeKernelConfiguration.h
similarity index 87%
rename from arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfiguration.h
rename to src/core/CL/gemm/native/CLGEMMNativeKernelConfiguration.h
index 7270a8e..aecf5a8 100644
--- a/arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfiguration.h
+++ b/src/core/CL/gemm/native/CLGEMMNativeKernelConfiguration.h
@@ -24,12 +24,12 @@
 #ifndef ARM_COMPUTE_CLGEMMNATIVEKERNELCONFIGURATION_H
 #define ARM_COMPUTE_CLGEMMNATIVEKERNELCONFIGURATION_H
 
-#include "arm_compute/core/CL/ICLGEMMKernelConfiguration.h"
-#include "arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.h"
-#include "arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.h"
-#include "arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.h"
+#include "src/core/CL/ICLGEMMKernelConfiguration.h"
+#include "src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.h"
+#include "src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.h"
+#include "src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.h"
 
-#include <memory>
+#include "support/MemorySupport.h"
 
 namespace arm_compute
 {
diff --git a/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.cpp b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.cpp
index 51b7fc7..4cc3d6a 100644
--- a/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.cpp
+++ b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.h"
+#include "src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
 #include "arm_compute/core/GPUTarget.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
 
 #include <map>
 #include <utility>
diff --git a/arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.h b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.h
similarity index 97%
rename from arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.h
rename to src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.h
index 1e49896..1e7432c 100644
--- a/arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.h
+++ b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationBifrost.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CLGEMMNATIVEKERNELCONFIGURATIONBIFROST_H
 #define ARM_COMPUTE_CLGEMMNATIVEKERNELCONFIGURATIONBIFROST_H
 
-#include "arm_compute/core/CL/ICLGEMMKernelConfiguration.h"
+#include "src/core/CL/ICLGEMMKernelConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.cpp b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.cpp
index 3e7c176..fd699a0 100644
--- a/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.cpp
+++ b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.h"
+#include "src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
 #include "arm_compute/core/GPUTarget.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
 
 #include <map>
 #include <utility>
diff --git a/arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.h b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.h
similarity index 96%
rename from arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.h
rename to src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.h
index 4cebfce..2f66717 100644
--- a/arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.h
+++ b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationMidgard.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CLGEMMNATIVEKERNELCONFIGURATIONMIDGARD_H
 #define ARM_COMPUTE_CLGEMMNATIVEKERNELCONFIGURATIONMIDGARD_H
 
-#include "arm_compute/core/CL/ICLGEMMKernelConfiguration.h"
+#include "src/core/CL/ICLGEMMKernelConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.cpp b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.cpp
index efc82fb..2c82340 100644
--- a/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.cpp
+++ b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.h"
+#include "src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
 #include "arm_compute/core/GPUTarget.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
 
 #include <map>
 #include <utility>
diff --git a/arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.h b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.h
similarity index 97%
rename from arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.h
rename to src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.h
index 07389ea..fb51b02 100644
--- a/arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.h
+++ b/src/core/CL/gemm/native/CLGEMMNativeKernelConfigurationValhall.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CLGEMMNATIVEKERNELCONFIGURATIONVALHALL_H
 #define ARM_COMPUTE_CLGEMMNATIVEKERNELCONFIGURATIONVALHALL_H
 
-#include "arm_compute/core/CL/ICLGEMMKernelConfiguration.h"
+#include "src/core/CL/ICLGEMMKernelConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfiguration.h b/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfiguration.h
similarity index 89%
rename from arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfiguration.h
rename to src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfiguration.h
index b953fd2..21ccf2d 100644
--- a/arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfiguration.h
+++ b/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfiguration.h
@@ -24,11 +24,11 @@
 #ifndef ARM_COMPUTE_CLGEMMRESHAPEDKERNELCONFIGURATION_H
 #define ARM_COMPUTE_CLGEMMRESHAPEDKERNELCONFIGURATION_H
 
-#include "arm_compute/core/CL/ICLGEMMKernelConfiguration.h"
-#include "arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.h"
-#include "arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.h"
+#include "src/core/CL/ICLGEMMKernelConfiguration.h"
+#include "src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.h"
+#include "src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.h"
 
-#include <memory>
+#include "support/MemorySupport.h"
 
 namespace arm_compute
 {
diff --git a/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.cpp b/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.cpp
index b5fc074..00c284f 100644
--- a/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.cpp
+++ b/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.cpp
@@ -21,15 +21,15 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.h"
+#include "src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
 #include "arm_compute/core/GPUTarget.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/TensorShape.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
 
 #include <map>
 #include <utility>
@@ -216,17 +216,17 @@
         {
             if(workload <= 790.39f)
             {
-                return configure_lhs_rhs_info(m,n,2,4,4,2,2,false,false,true,false,false);
+                return configure_lhs_rhs_info(m, n, 2, 4, 4, 2, 2, false, false, true, false, false);
             }
             else
             {
                 if(workload <= 982.39f)
                 {
-                    return configure_lhs_rhs_info(m,n,4,2,4,4,4,false,false,true,false,false);
+                    return configure_lhs_rhs_info(m, n, 4, 2, 4, 4, 4, false, false, true, false, false);
                 }
                 else
                 {
-                    return configure_lhs_rhs_info(m,n,2,4,4,2,1,false,true,true,false,false);
+                    return configure_lhs_rhs_info(m, n, 2, 4, 4, 2, 1, false, true, true, false, false);
                 }
             }
         }
@@ -236,16 +236,16 @@
             {
                 if(r_mn <= 0.11f)
                 {
-                    return configure_lhs_rhs_info(m,n,2,4,4,2,2,false,false,true,false,false);
+                    return configure_lhs_rhs_info(m, n, 2, 4, 4, 2, 2, false, false, true, false, false);
                 }
                 else
                 {
-                    return configure_lhs_rhs_info(m,n,4,4,4,4,4,false,true,true,false,false);
+                    return configure_lhs_rhs_info(m, n, 4, 4, 4, 4, 4, false, true, true, false, false);
                 }
             }
             else
             {
-                return configure_lhs_rhs_info(m,n,2,4,4,2,2,false,false,true,false,false);
+                return configure_lhs_rhs_info(m, n, 2, 4, 4, 2, 2, false, false, true, false, false);
             }
         }
     }
@@ -257,22 +257,22 @@
             {
                 if(m <= 64.5)
                 {
-                    return configure_lhs_rhs_info(m,n,4,4,4,2,4,true,false,true,false,false);
+                    return configure_lhs_rhs_info(m, n, 4, 4, 4, 2, 4, true, false, true, false, false);
                 }
                 else
                 {
-                    return configure_lhs_rhs_info(m,n,4,4,4,2,2,false,true,true,false,false);
+                    return configure_lhs_rhs_info(m, n, 4, 4, 4, 2, 2, false, true, true, false, false);
                 }
             }
             else
             {
                 if(r_mn <= 1.09f)
                 {
-                    return configure_lhs_rhs_info(m,n,4,4,4,4,4,false,true,true,false,false);
+                    return configure_lhs_rhs_info(m, n, 4, 4, 4, 4, 4, false, true, true, false, false);
                 }
                 else
                 {
-                    return configure_lhs_rhs_info(m,n,4,4,4,2,2,true,true,true,false,false);
+                    return configure_lhs_rhs_info(m, n, 4, 4, 4, 2, 2, true, true, true, false, false);
                 }
             }
         }
@@ -280,17 +280,17 @@
         {
             if(m <= 43)
             {
-                return configure_lhs_rhs_info(m,n,4,4,4,2,4,true,false,true,false,false);
+                return configure_lhs_rhs_info(m, n, 4, 4, 4, 2, 4, true, false, true, false, false);
             }
             else
             {
                 if(workload <= 26364.79f)
                 {
-                    return configure_lhs_rhs_info(m,n,4,4,4,2,2,false,true,true,false,false);
+                    return configure_lhs_rhs_info(m, n, 4, 4, 4, 2, 2, false, true, true, false, false);
                 }
                 else
                 {
-                    return configure_lhs_rhs_info(m,n,4,4,4,4,4,false,true,true,false,false);
+                    return configure_lhs_rhs_info(m, n, 4, 4, 4, 4, 4, false, true, true, false, false);
                 }
             }
         }
diff --git a/arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.h b/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.h
similarity index 97%
rename from arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.h
rename to src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.h
index 4df2784..e3b62ce 100644
--- a/arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.h
+++ b/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationBifrost.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CLGEMMRESHAPEDKERNELCONFIGURATIONBIFROST_H
 #define ARM_COMPUTE_CLGEMMRESHAPEDKERNELCONFIGURATIONBIFROST_H
 
-#include "arm_compute/core/CL/ICLGEMMKernelConfiguration.h"
+#include "src/core/CL/ICLGEMMKernelConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.cpp b/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.cpp
index 0c09f50..519e903 100644
--- a/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.cpp
+++ b/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.h"
+#include "src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
 #include "arm_compute/core/GPUTarget.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
 
 #include <map>
 #include <utility>
diff --git a/arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.h b/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.h
similarity index 97%
rename from arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.h
rename to src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.h
index 7a617e0..5f7e701 100644
--- a/arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.h
+++ b/src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfigurationValhall.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CLGEMMRESHAPEDKERNELCONFIGURATIONVALHALL_H
 #define ARM_COMPUTE_CLGEMMRESHAPEDKERNELCONFIGURATIONVALHALL_H
 
-#include "arm_compute/core/CL/ICLGEMMKernelConfiguration.h"
+#include "src/core/CL/ICLGEMMKernelConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfiguration.h b/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfiguration.h
similarity index 88%
rename from arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfiguration.h
rename to src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfiguration.h
index 6d5ce88..4efe28c 100644
--- a/arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfiguration.h
+++ b/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfiguration.h
@@ -24,11 +24,11 @@
 #ifndef ARM_COMPUTE_CLGEMMRESHAPEDONLYRHSKERNELCONFIGURATION_H
 #define ARM_COMPUTE_CLGEMMRESHAPEDONLYRHSKERNELCONFIGURATION_H
 
-#include "arm_compute/core/CL/ICLGEMMKernelConfiguration.h"
-#include "arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.h"
-#include "arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.h"
+#include "src/core/CL/ICLGEMMKernelConfiguration.h"
+#include "src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.h"
+#include "src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.h"
 
-#include <memory>
+#include "support/MemorySupport.h"
 
 namespace arm_compute
 {
diff --git a/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.cpp b/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.cpp
index 16eabf0..0a0fc5d 100644
--- a/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.cpp
+++ b/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.cpp
@@ -21,15 +21,15 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.h"
+#include "src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
 #include "arm_compute/core/GPUTarget.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/TensorShape.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
 
 #include <map>
 #include <utility>
@@ -122,13 +122,13 @@
 
     if(m == 1)
     {
-        if ( n <= 2548 )
+        if(n <= 2548)
         {
-            return configure_lhs_rhs_info(m,n,1,2,16,1,4,false,true,false,true,false);
+            return configure_lhs_rhs_info(m, n, 1, 2, 16, 1, 4, false, true, false, true, false);
         }
         else
         {
-            return configure_lhs_rhs_info(m,n,1,4,16,1,8,false,true,false,true,false);
+            return configure_lhs_rhs_info(m, n, 1, 4, 16, 1, 8, false, true, false, true, false);
         }
     }
     else
@@ -251,7 +251,7 @@
 
     if(m == 1)
     {
-        return configure_lhs_rhs_info(m,n,1,2,16,1,32,false,true,false,true,false);
+        return configure_lhs_rhs_info(m, n, 1, 2, 16, 1, 32, false, true, false, true, false);
     }
     else
     {
diff --git a/arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.h b/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.h
similarity index 97%
rename from arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.h
rename to src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.h
index 346bfd7..618dbd9 100644
--- a/arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.h
+++ b/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationBifrost.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CLGEMMRESHAPEDONLYRHSKERNELCONFIGURATIONBIFROST_H
 #define ARM_COMPUTE_CLGEMMRESHAPEDONLYRHSKERNELCONFIGURATIONBIFROST_H
 
-#include "arm_compute/core/CL/ICLGEMMKernelConfiguration.h"
+#include "src/core/CL/ICLGEMMKernelConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.cpp b/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.cpp
index 23bf02c..f7939d2 100644
--- a/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.cpp
+++ b/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.cpp
@@ -21,15 +21,15 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.h"
+#include "src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
 #include "arm_compute/core/GPUTarget.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/TensorShape.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
 
 #include <map>
 #include <utility>
diff --git a/arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.h b/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.h
similarity index 97%
rename from arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.h
rename to src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.h
index 2162baf..b928992 100644
--- a/arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.h
+++ b/src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfigurationValhall.h
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_CLGEMMRESHAPEDONLYRHSKERNELCONFIGURATIONVALHALL_H
 #define ARM_COMPUTE_CLGEMMRESHAPEDONLYRHSKERNELCONFIGURATIONVALHALL_H
 
-#include "arm_compute/core/CL/ICLGEMMKernelConfiguration.h"
+#include "src/core/CL/ICLGEMMKernelConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/core/CL/kernels/CLAbsoluteDifferenceKernel.cpp b/src/core/CL/kernels/CLAbsoluteDifferenceKernel.cpp
index 9deb165..29745be 100644
--- a/src/core/CL/kernels/CLAbsoluteDifferenceKernel.cpp
+++ b/src/core/CL/kernels/CLAbsoluteDifferenceKernel.cpp
@@ -27,12 +27,9 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <set>
 #include <string>
diff --git a/src/core/CL/kernels/CLActivationLayerKernel.cpp b/src/core/CL/kernels/CLActivationLayerKernel.cpp
index 5a9434e..f0e3047 100644
--- a/src/core/CL/kernels/CLActivationLayerKernel.cpp
+++ b/src/core/CL/kernels/CLActivationLayerKernel.cpp
@@ -25,14 +25,14 @@
 
 #include "arm_compute/core/CL/CLCoreRuntimeContext.h"
 #include "arm_compute/core/CL/CLHelpers.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/helpers/float_ops.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/Cast.h"
+
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLArgMinMaxLayerKernel.cpp b/src/core/CL/kernels/CLArgMinMaxLayerKernel.cpp
index b78ac27..b5a801a 100644
--- a/src/core/CL/kernels/CLArgMinMaxLayerKernel.cpp
+++ b/src/core/CL/kernels/CLArgMinMaxLayerKernel.cpp
@@ -23,16 +23,17 @@
  */
 #include "arm_compute/core/CL/kernels/CLArgMinMaxLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLBatchConcatenateLayerKernel.cpp b/src/core/CL/kernels/CLBatchConcatenateLayerKernel.cpp
index dd9c234..7a8c9ad 100644
--- a/src/core/CL/kernels/CLBatchConcatenateLayerKernel.cpp
+++ b/src/core/CL/kernels/CLBatchConcatenateLayerKernel.cpp
@@ -25,12 +25,12 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/Cast.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLBatchNormalizationLayerKernel.cpp b/src/core/CL/kernels/CLBatchNormalizationLayerKernel.cpp
index 1c1df6c..09b668d 100644
--- a/src/core/CL/kernels/CLBatchNormalizationLayerKernel.cpp
+++ b/src/core/CL/kernels/CLBatchNormalizationLayerKernel.cpp
@@ -25,12 +25,13 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLBatchToSpaceLayerKernel.cpp b/src/core/CL/kernels/CLBatchToSpaceLayerKernel.cpp
index c74f7e0..6eb22c9 100644
--- a/src/core/CL/kernels/CLBatchToSpaceLayerKernel.cpp
+++ b/src/core/CL/kernels/CLBatchToSpaceLayerKernel.cpp
@@ -24,9 +24,11 @@
 #include "arm_compute/core/CL/kernels/CLBatchToSpaceLayerKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute::misc::shape_calculator;
diff --git a/src/core/CL/kernels/CLBitwiseAndKernel.cpp b/src/core/CL/kernels/CLBitwiseAndKernel.cpp
index 44378c8..53a438d 100644
--- a/src/core/CL/kernels/CLBitwiseAndKernel.cpp
+++ b/src/core/CL/kernels/CLBitwiseAndKernel.cpp
@@ -27,9 +27,8 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute;
 
diff --git a/src/core/CL/kernels/CLBitwiseOrKernel.cpp b/src/core/CL/kernels/CLBitwiseOrKernel.cpp
index 77c48e6..0e2e5d4 100644
--- a/src/core/CL/kernels/CLBitwiseOrKernel.cpp
+++ b/src/core/CL/kernels/CLBitwiseOrKernel.cpp
@@ -27,9 +27,8 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute;
 
diff --git a/src/core/CL/kernels/CLBitwiseXorKernel.cpp b/src/core/CL/kernels/CLBitwiseXorKernel.cpp
index a15305e..65b17c0 100644
--- a/src/core/CL/kernels/CLBitwiseXorKernel.cpp
+++ b/src/core/CL/kernels/CLBitwiseXorKernel.cpp
@@ -27,9 +27,8 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute;
 
diff --git a/src/core/CL/kernels/CLBoundingBoxTransformKernel.cpp b/src/core/CL/kernels/CLBoundingBoxTransformKernel.cpp
index 95ea3d7..b8c0d2f 100644
--- a/src/core/CL/kernels/CLBoundingBoxTransformKernel.cpp
+++ b/src/core/CL/kernels/CLBoundingBoxTransformKernel.cpp
@@ -23,17 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLBoundingBoxTransformKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLArray.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLBox3x3Kernel.cpp b/src/core/CL/kernels/CLBox3x3Kernel.cpp
index 7916dce..2f6c09d 100644
--- a/src/core/CL/kernels/CLBox3x3Kernel.cpp
+++ b/src/core/CL/kernels/CLBox3x3Kernel.cpp
@@ -26,9 +26,8 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/IAccessWindow.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <set>
 #include <string>
diff --git a/src/core/CL/kernels/CLCannyEdgeKernel.cpp b/src/core/CL/kernels/CLCannyEdgeKernel.cpp
index b8a5365..c76ec67 100644
--- a/src/core/CL/kernels/CLCannyEdgeKernel.cpp
+++ b/src/core/CL/kernels/CLCannyEdgeKernel.cpp
@@ -29,6 +29,7 @@
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/CL/kernels/CLChannelCombineKernel.cpp b/src/core/CL/kernels/CLChannelCombineKernel.cpp
index b0e5111..d574f35 100644
--- a/src/core/CL/kernels/CLChannelCombineKernel.cpp
+++ b/src/core/CL/kernels/CLChannelCombineKernel.cpp
@@ -27,14 +27,12 @@
 #include "arm_compute/core/CL/ICLMultiImage.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/MultiImageInfo.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <set>
 #include <string>
diff --git a/src/core/CL/kernels/CLChannelExtractKernel.cpp b/src/core/CL/kernels/CLChannelExtractKernel.cpp
index 13ae8f5..7911b94 100644
--- a/src/core/CL/kernels/CLChannelExtractKernel.cpp
+++ b/src/core/CL/kernels/CLChannelExtractKernel.cpp
@@ -28,14 +28,13 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Coordinates.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/MultiImageInfo.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <set>
 #include <string>
diff --git a/src/core/CL/kernels/CLChannelShuffleLayerKernel.cpp b/src/core/CL/kernels/CLChannelShuffleLayerKernel.cpp
index ad000ba..301a762 100644
--- a/src/core/CL/kernels/CLChannelShuffleLayerKernel.cpp
+++ b/src/core/CL/kernels/CLChannelShuffleLayerKernel.cpp
@@ -25,12 +25,12 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLCol2ImKernel.cpp b/src/core/CL/kernels/CLCol2ImKernel.cpp
index 4050b24..3dc007d 100644
--- a/src/core/CL/kernels/CLCol2ImKernel.cpp
+++ b/src/core/CL/kernels/CLCol2ImKernel.cpp
@@ -25,13 +25,13 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cmath>
diff --git a/src/core/CL/kernels/CLColorConvertKernel.cpp b/src/core/CL/kernels/CLColorConvertKernel.cpp
index e14b871..0f82d87 100644
--- a/src/core/CL/kernels/CLColorConvertKernel.cpp
+++ b/src/core/CL/kernels/CLColorConvertKernel.cpp
@@ -27,14 +27,12 @@
 #include "arm_compute/core/CL/ICLMultiImage.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/MultiImageInfo.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <sstream>
diff --git a/src/core/CL/kernels/CLComparisonKernel.cpp b/src/core/CL/kernels/CLComparisonKernel.cpp
index 5bb1d56..2b72946 100644
--- a/src/core/CL/kernels/CLComparisonKernel.cpp
+++ b/src/core/CL/kernels/CLComparisonKernel.cpp
@@ -24,8 +24,10 @@
 #include "arm_compute/core/CL/kernels/CLComparisonKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <map>
diff --git a/src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp b/src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp
index 7c61146..b4e42bf 100644
--- a/src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp
+++ b/src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp
@@ -25,10 +25,11 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLConvolutionKernel.cpp b/src/core/CL/kernels/CLConvolutionKernel.cpp
index ca07e68..48b185f 100644
--- a/src/core/CL/kernels/CLConvolutionKernel.cpp
+++ b/src/core/CL/kernels/CLConvolutionKernel.cpp
@@ -31,9 +31,9 @@
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLCopyKernel.cpp b/src/core/CL/kernels/CLCopyKernel.cpp
index 37c3241..0b7e9af 100644
--- a/src/core/CL/kernels/CLCopyKernel.cpp
+++ b/src/core/CL/kernels/CLCopyKernel.cpp
@@ -23,15 +23,15 @@
  */
 #include "arm_compute/core/CL/kernels/CLCopyKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLCropKernel.cpp b/src/core/CL/kernels/CLCropKernel.cpp
index f828162..2c99d46 100644
--- a/src/core/CL/kernels/CLCropKernel.cpp
+++ b/src/core/CL/kernels/CLCropKernel.cpp
@@ -26,16 +26,11 @@
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/WindowHelpers.h"
 
-#include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
-#include "arm_compute/core/utils/helpers/bit_ops.h"
-#include "arm_compute/core/utils/helpers/tensor_transform.h"
-#include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "support/StringSupport.h"
 
 #include <map>
diff --git a/src/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.cpp b/src/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.cpp
index e8f12d5..9ba3dc3 100644
--- a/src/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.cpp
+++ b/src/core/CL/kernels/CLDeconvolutionLayerUpsampleKernel.cpp
@@ -25,12 +25,11 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/CL/kernels/CLDeconvolutionReshapeOutputKernel.cpp b/src/core/CL/kernels/CLDeconvolutionReshapeOutputKernel.cpp
index 6973034..1514d90 100644
--- a/src/core/CL/kernels/CLDeconvolutionReshapeOutputKernel.cpp
+++ b/src/core/CL/kernels/CLDeconvolutionReshapeOutputKernel.cpp
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLDepthConcatenateLayerKernel.cpp b/src/core/CL/kernels/CLDepthConcatenateLayerKernel.cpp
index 87067cf..cb5d727 100644
--- a/src/core/CL/kernels/CLDepthConcatenateLayerKernel.cpp
+++ b/src/core/CL/kernels/CLDepthConcatenateLayerKernel.cpp
@@ -25,12 +25,12 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/Cast.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLDepthConvertLayerKernel.cpp b/src/core/CL/kernels/CLDepthConvertLayerKernel.cpp
index 11297e7..24f638f 100644
--- a/src/core/CL/kernels/CLDepthConvertLayerKernel.cpp
+++ b/src/core/CL/kernels/CLDepthConvertLayerKernel.cpp
@@ -25,13 +25,13 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
 #include "support/StringSupport.h"
 
 #include <cstddef>
diff --git a/src/core/CL/kernels/CLDepthToSpaceLayerKernel.cpp b/src/core/CL/kernels/CLDepthToSpaceLayerKernel.cpp
index e6f909e..50bb3b8 100644
--- a/src/core/CL/kernels/CLDepthToSpaceLayerKernel.cpp
+++ b/src/core/CL/kernels/CLDepthToSpaceLayerKernel.cpp
@@ -24,9 +24,11 @@
 #include "arm_compute/core/CL/kernels/CLDepthToSpaceLayerKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute::misc::shape_calculator;
diff --git a/src/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.cpp b/src/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.cpp
index 066e9a5..7958230 100644
--- a/src/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.cpp
+++ b/src/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.cpp
@@ -23,19 +23,19 @@
  */
 #include "arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NCHWKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.cpp b/src/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.cpp
index 0930fee..5a0d2d0 100644
--- a/src/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.cpp
+++ b/src/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.cpp
@@ -23,19 +23,19 @@
  */
 #include "arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.cpp b/src/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.cpp
index a538ab5..5a3a0ec 100644
--- a/src/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.cpp
+++ b/src/core/CL/kernels/CLDepthwiseConvolutionLayerNativeKernel.cpp
@@ -25,17 +25,16 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLDepthwiseConvolutionLayerReshapeWeightsKernel.cpp b/src/core/CL/kernels/CLDepthwiseConvolutionLayerReshapeWeightsKernel.cpp
index 07f25a8..0ff3c52 100644
--- a/src/core/CL/kernels/CLDepthwiseConvolutionLayerReshapeWeightsKernel.cpp
+++ b/src/core/CL/kernels/CLDepthwiseConvolutionLayerReshapeWeightsKernel.cpp
@@ -23,19 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayerReshapeWeightsKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
-#include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLDequantizationLayerKernel.cpp b/src/core/CL/kernels/CLDequantizationLayerKernel.cpp
index 72eac85..e653c59 100644
--- a/src/core/CL/kernels/CLDequantizationLayerKernel.cpp
+++ b/src/core/CL/kernels/CLDequantizationLayerKernel.cpp
@@ -23,15 +23,16 @@
  */
 #include "arm_compute/core/CL/kernels/CLDequantizationLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLDerivativeKernel.cpp b/src/core/CL/kernels/CLDerivativeKernel.cpp
index ab5f9da..659a7cb 100644
--- a/src/core/CL/kernels/CLDerivativeKernel.cpp
+++ b/src/core/CL/kernels/CLDerivativeKernel.cpp
@@ -26,11 +26,9 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLDilateKernel.cpp b/src/core/CL/kernels/CLDilateKernel.cpp
index ae94831..1e59c34 100644
--- a/src/core/CL/kernels/CLDilateKernel.cpp
+++ b/src/core/CL/kernels/CLDilateKernel.cpp
@@ -26,8 +26,8 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute;
 
diff --git a/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp b/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp
index d5d808a..161b221 100644
--- a/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp
+++ b/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp
@@ -23,19 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLDirectConvolutionLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/ITensor.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLElementWiseUnaryLayerKernel.cpp b/src/core/CL/kernels/CLElementWiseUnaryLayerKernel.cpp
index c8c7fb0..c7f9df0 100644
--- a/src/core/CL/kernels/CLElementWiseUnaryLayerKernel.cpp
+++ b/src/core/CL/kernels/CLElementWiseUnaryLayerKernel.cpp
@@ -24,9 +24,10 @@
 #include "arm_compute/core/CL/kernels/CLElementWiseUnaryLayerKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/Cast.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLElementwiseOperationKernel.cpp b/src/core/CL/kernels/CLElementwiseOperationKernel.cpp
index ec33500..f0712f5 100644
--- a/src/core/CL/kernels/CLElementwiseOperationKernel.cpp
+++ b/src/core/CL/kernels/CLElementwiseOperationKernel.cpp
@@ -24,9 +24,11 @@
 #include "arm_compute/core/CL/kernels/CLElementwiseOperationKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/Cast.h"
 #include "support/StringSupport.h"
 #include <map>
 
diff --git a/src/core/CL/kernels/CLErodeKernel.cpp b/src/core/CL/kernels/CLErodeKernel.cpp
index a5eb79f..29a3297 100644
--- a/src/core/CL/kernels/CLErodeKernel.cpp
+++ b/src/core/CL/kernels/CLErodeKernel.cpp
@@ -26,8 +26,8 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute;
 
diff --git a/src/core/CL/kernels/CLFFTDigitReverseKernel.cpp b/src/core/CL/kernels/CLFFTDigitReverseKernel.cpp
index 30bca2f..0478f55 100644
--- a/src/core/CL/kernels/CLFFTDigitReverseKernel.cpp
+++ b/src/core/CL/kernels/CLFFTDigitReverseKernel.cpp
@@ -24,11 +24,11 @@
 #include "arm_compute/core/CL/kernels/CLFFTDigitReverseKernel.h"
 
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLFFTRadixStageKernel.cpp b/src/core/CL/kernels/CLFFTRadixStageKernel.cpp
index 6c36338..7b17a22 100644
--- a/src/core/CL/kernels/CLFFTRadixStageKernel.cpp
+++ b/src/core/CL/kernels/CLFFTRadixStageKernel.cpp
@@ -25,12 +25,12 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cmath>
diff --git a/src/core/CL/kernels/CLFFTScaleKernel.cpp b/src/core/CL/kernels/CLFFTScaleKernel.cpp
index ac5f2b3..49fcbb6 100644
--- a/src/core/CL/kernels/CLFFTScaleKernel.cpp
+++ b/src/core/CL/kernels/CLFFTScaleKernel.cpp
@@ -24,11 +24,11 @@
 #include "arm_compute/core/CL/kernels/CLFFTScaleKernel.h"
 
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLFastCornersKernel.cpp b/src/core/CL/kernels/CLFastCornersKernel.cpp
index e71b472..ebdfd27 100644
--- a/src/core/CL/kernels/CLFastCornersKernel.cpp
+++ b/src/core/CL/kernels/CLFastCornersKernel.cpp
@@ -26,11 +26,9 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLFillBorderKernel.cpp b/src/core/CL/kernels/CLFillBorderKernel.cpp
index 1ea654b..e92619a 100644
--- a/src/core/CL/kernels/CLFillBorderKernel.cpp
+++ b/src/core/CL/kernels/CLFillBorderKernel.cpp
@@ -27,13 +27,11 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/Cast.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLFlattenLayerKernel.cpp b/src/core/CL/kernels/CLFlattenLayerKernel.cpp
index 6bd1149..dc1d338 100644
--- a/src/core/CL/kernels/CLFlattenLayerKernel.cpp
+++ b/src/core/CL/kernels/CLFlattenLayerKernel.cpp
@@ -23,9 +23,10 @@
  */
 #include "arm_compute/core/CL/kernels/CLFlattenLayerKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute::misc::shape_calculator;
diff --git a/src/core/CL/kernels/CLFloorKernel.cpp b/src/core/CL/kernels/CLFloorKernel.cpp
index 09f5f61..8884f3f 100644
--- a/src/core/CL/kernels/CLFloorKernel.cpp
+++ b/src/core/CL/kernels/CLFloorKernel.cpp
@@ -25,14 +25,14 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLFuseBatchNormalizationKernel.cpp b/src/core/CL/kernels/CLFuseBatchNormalizationKernel.cpp
index b582295..61e2b27 100644
--- a/src/core/CL/kernels/CLFuseBatchNormalizationKernel.cpp
+++ b/src/core/CL/kernels/CLFuseBatchNormalizationKernel.cpp
@@ -25,12 +25,13 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyNativeKernel.cpp b/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyNativeKernel.cpp
index d30a9e5..cc98845 100644
--- a/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyNativeKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyNativeKernel.cpp
@@ -23,19 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMLowpMatrixMultiplyNativeKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cstddef>
@@ -137,7 +136,7 @@
     num_elems_processed_per_iteration_x = rhs_info.n0;
     num_elems_processed_per_iteration_y = lhs_info.m0;
 
-    win     = calculate_max_window(tmp_info, Steps(num_elems_processed_per_iteration_x, num_elems_processed_per_iteration_y));
+    win = calculate_max_window(tmp_info, Steps(num_elems_processed_per_iteration_x, num_elems_processed_per_iteration_y));
     output->set_valid_region(ValidRegion(Coordinates(), output->tensor_shape()));
 
     // Collapse along the Z direction
diff --git a/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedKernel.cpp b/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedKernel.cpp
index 56b92a3..5469c89 100644
--- a/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedKernel.cpp
@@ -23,19 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cstddef>
diff --git a/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedOnlyRHSKernel.cpp b/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedOnlyRHSKernel.cpp
index 4770329..4a3ac2d 100644
--- a/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedOnlyRHSKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedOnlyRHSKernel.cpp
@@ -23,19 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMLowpMatrixMultiplyReshapedOnlyRHSKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cstddef>
diff --git a/src/core/CL/kernels/CLGEMMLowpOffsetContributionKernel.cpp b/src/core/CL/kernels/CLGEMMLowpOffsetContributionKernel.cpp
index 6ef9fd2..7ab96e5 100644
--- a/src/core/CL/kernels/CLGEMMLowpOffsetContributionKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMLowpOffsetContributionKernel.cpp
@@ -23,15 +23,13 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMLowpOffsetContributionKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cstddef>
@@ -148,7 +146,8 @@
     configure(CLKernelLibrary::get().get_compile_context(), mm_result, vector_sum_col, vector_sum_row, bias, k, a_offset, b_offset);
 }
 
-void CLGEMMLowpOffsetContributionKernel::configure(const CLCompileContext &compile_context, ICLTensor *mm_result, const ICLTensor *vector_sum_col, const ICLTensor *vector_sum_row, const ICLTensor *bias,
+void CLGEMMLowpOffsetContributionKernel::configure(const CLCompileContext &compile_context, ICLTensor *mm_result, const ICLTensor *vector_sum_col, const ICLTensor *vector_sum_row,
+                                                   const ICLTensor *bias,
                                                    int32_t k, int32_t a_offset,
                                                    int32_t b_offset)
 {
diff --git a/src/core/CL/kernels/CLGEMMLowpOffsetContributionOutputStageKernel.cpp b/src/core/CL/kernels/CLGEMMLowpOffsetContributionOutputStageKernel.cpp
index 6d3aa6f..85285d6 100644
--- a/src/core/CL/kernels/CLGEMMLowpOffsetContributionOutputStageKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMLowpOffsetContributionOutputStageKernel.cpp
@@ -23,16 +23,15 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMLowpOffsetContributionOutputStageKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cstddef>
diff --git a/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp b/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp
index eae6641..ab1b5a2 100644
--- a/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFixedPointKernel.cpp
@@ -25,14 +25,14 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel.cpp b/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel.cpp
index 430a84c..ad5bac0 100644
--- a/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleByFloatKernel.cpp
@@ -25,14 +25,14 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleKernel.cpp b/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleKernel.cpp
index 79888cd..8e4b291 100644
--- a/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ScaleKernel.cpp
@@ -25,12 +25,12 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp b/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp
index 31a97ca..826b265 100644
--- a/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp
@@ -23,10 +23,12 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMLowpReductionKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/KernelDescriptors.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLGEMMMatrixMultiplyKernel.cpp b/src/core/CL/kernels/CLGEMMMatrixMultiplyKernel.cpp
index c2dd92c..aa69ed0 100644
--- a/src/core/CL/kernels/CLGEMMMatrixMultiplyKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMMatrixMultiplyKernel.cpp
@@ -23,20 +23,19 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/helpers/float_ops.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/utils/helpers/float_ops.h"
 #include "support/StringSupport.h"
 
 #include <set>
@@ -310,7 +309,8 @@
     configure(CLKernelLibrary::get().get_compile_context(), input0, input1, input2, output, alpha, beta, is_interleaved_transposed, reshape_info, fp_mixed_precision, activation_info);
 }
 
-void CLGEMMMatrixMultiplyKernel::configure(const CLCompileContext &compile_context, const ICLTensor *input0, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output, float alpha, float beta,
+void CLGEMMMatrixMultiplyKernel::configure(const CLCompileContext &compile_context, const ICLTensor *input0, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output, float alpha,
+                                           float beta,
                                            bool is_interleaved_transposed, const GEMMReshapeInfo &reshape_info, bool fp_mixed_precision, const ActivationLayerInfo &activation_info)
 {
     ARM_COMPUTE_ERROR_ON_NULLPTR(input0, input1, output);
diff --git a/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp b/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp
index da57aa4..cea147b 100644
--- a/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp
@@ -23,20 +23,19 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/helpers/float_ops.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/utils/helpers/float_ops.h"
 #include "support/StringSupport.h"
 
 #include <cstddef>
diff --git a/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedKernel.cpp b/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedKernel.cpp
index b0f0e8a..eaf5708 100644
--- a/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedKernel.cpp
@@ -23,23 +23,22 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyReshapedKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/helpers/float_ops.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
 #include "src/core/CL/CLUtils.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/utils/helpers/float_ops.h"
 #include "support/StringSupport.h"
 
 #include <cstddef>
diff --git a/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.cpp b/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.cpp
index 0ae30ed..912c763 100644
--- a/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.cpp
@@ -23,17 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/utils/helpers/float_ops.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
 #include "src/core/CL/CLUtils.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/utils/helpers/float_ops.h"
 #include "support/StringSupport.h"
 
 #include <tuple>
diff --git a/src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp b/src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp
index f523845..04aa061 100644
--- a/src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp
@@ -23,15 +23,14 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLGEMMReshapeLHSMatrixKernel.cpp b/src/core/CL/kernels/CLGEMMReshapeLHSMatrixKernel.cpp
index 156a657..f2ad677 100644
--- a/src/core/CL/kernels/CLGEMMReshapeLHSMatrixKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMReshapeLHSMatrixKernel.cpp
@@ -23,19 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMReshapeLHSMatrixKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLGEMMReshapeRHSMatrixKernel.cpp b/src/core/CL/kernels/CLGEMMReshapeRHSMatrixKernel.cpp
index ce29464..d94e834 100644
--- a/src/core/CL/kernels/CLGEMMReshapeRHSMatrixKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMReshapeRHSMatrixKernel.cpp
@@ -23,20 +23,19 @@
  */
 #include "arm_compute/core/CL/kernels/CLGEMMReshapeRHSMatrixKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLGatherKernel.cpp b/src/core/CL/kernels/CLGatherKernel.cpp
index 57759fc..a8508be 100644
--- a/src/core/CL/kernels/CLGatherKernel.cpp
+++ b/src/core/CL/kernels/CLGatherKernel.cpp
@@ -23,8 +23,9 @@
  */
 #include "arm_compute/core/CL/kernels/CLGatherKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <string>
diff --git a/src/core/CL/kernels/CLGaussian3x3Kernel.cpp b/src/core/CL/kernels/CLGaussian3x3Kernel.cpp
index 08e7e27..c9ed1ac 100644
--- a/src/core/CL/kernels/CLGaussian3x3Kernel.cpp
+++ b/src/core/CL/kernels/CLGaussian3x3Kernel.cpp
@@ -26,8 +26,8 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <set>
 #include <string>
diff --git a/src/core/CL/kernels/CLGaussianPyramidKernel.cpp b/src/core/CL/kernels/CLGaussianPyramidKernel.cpp
index 0e20187..2686e8b 100644
--- a/src/core/CL/kernels/CLGaussianPyramidKernel.cpp
+++ b/src/core/CL/kernels/CLGaussianPyramidKernel.cpp
@@ -28,6 +28,7 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/CL/kernels/CLGenerateProposalsLayerKernel.cpp b/src/core/CL/kernels/CLGenerateProposalsLayerKernel.cpp
index 3108ad8..a2fcbba 100644
--- a/src/core/CL/kernels/CLGenerateProposalsLayerKernel.cpp
+++ b/src/core/CL/kernels/CLGenerateProposalsLayerKernel.cpp
@@ -23,17 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLGenerateProposalsLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLArray.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLHOGDescriptorKernel.cpp b/src/core/CL/kernels/CLHOGDescriptorKernel.cpp
index 7f618b2..eaf5ea4 100644
--- a/src/core/CL/kernels/CLHOGDescriptorKernel.cpp
+++ b/src/core/CL/kernels/CLHOGDescriptorKernel.cpp
@@ -27,12 +27,10 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLHOGDetectorKernel.cpp b/src/core/CL/kernels/CLHOGDetectorKernel.cpp
index fbd2208..6e14996 100644
--- a/src/core/CL/kernels/CLHOGDetectorKernel.cpp
+++ b/src/core/CL/kernels/CLHOGDetectorKernel.cpp
@@ -27,12 +27,10 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLHOG.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/CL/kernels/CLHarrisCornersKernel.cpp b/src/core/CL/kernels/CLHarrisCornersKernel.cpp
index 08e670f..19c4e57 100644
--- a/src/core/CL/kernels/CLHarrisCornersKernel.cpp
+++ b/src/core/CL/kernels/CLHarrisCornersKernel.cpp
@@ -23,17 +23,16 @@
  */
 #include "arm_compute/core/CL/kernels/CLHarrisCornersKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLHeightConcatenateLayerKernel.cpp b/src/core/CL/kernels/CLHeightConcatenateLayerKernel.cpp
index 22b2cfc..1ae2599 100644
--- a/src/core/CL/kernels/CLHeightConcatenateLayerKernel.cpp
+++ b/src/core/CL/kernels/CLHeightConcatenateLayerKernel.cpp
@@ -25,13 +25,13 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/Cast.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLHistogramKernel.cpp b/src/core/CL/kernels/CLHistogramKernel.cpp
index b8a4e86..a85429c 100644
--- a/src/core/CL/kernels/CLHistogramKernel.cpp
+++ b/src/core/CL/kernels/CLHistogramKernel.cpp
@@ -27,12 +27,10 @@
 #include "arm_compute/core/CL/ICLDistribution1D.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cstring>
diff --git a/src/core/CL/kernels/CLIm2ColKernel.cpp b/src/core/CL/kernels/CLIm2ColKernel.cpp
index c94e313..76490f8 100644
--- a/src/core/CL/kernels/CLIm2ColKernel.cpp
+++ b/src/core/CL/kernels/CLIm2ColKernel.cpp
@@ -23,18 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLIm2ColKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cmath>
diff --git a/src/core/CL/kernels/CLInstanceNormalizationLayerKernel.cpp b/src/core/CL/kernels/CLInstanceNormalizationLayerKernel.cpp
index 2ad5233..e97b856 100644
--- a/src/core/CL/kernels/CLInstanceNormalizationLayerKernel.cpp
+++ b/src/core/CL/kernels/CLInstanceNormalizationLayerKernel.cpp
@@ -25,12 +25,13 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLIntegralImageKernel.cpp b/src/core/CL/kernels/CLIntegralImageKernel.cpp
index aff4bd9..82f6da8 100644
--- a/src/core/CL/kernels/CLIntegralImageKernel.cpp
+++ b/src/core/CL/kernels/CLIntegralImageKernel.cpp
@@ -28,9 +28,8 @@
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cstddef>
diff --git a/src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp b/src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp
index a68d8db..9936e29 100644
--- a/src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp
+++ b/src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp
@@ -25,13 +25,14 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLLKTrackerKernel.cpp b/src/core/CL/kernels/CLLKTrackerKernel.cpp
index fae5fe2..0fa2e70 100644
--- a/src/core/CL/kernels/CLLKTrackerKernel.cpp
+++ b/src/core/CL/kernels/CLLKTrackerKernel.cpp
@@ -23,16 +23,15 @@
  */
 #include "arm_compute/core/CL/kernels/CLLKTrackerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLArray.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Coordinates.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <cmath>
 
diff --git a/src/core/CL/kernels/CLLocallyConnectedMatrixMultiplyKernel.cpp b/src/core/CL/kernels/CLLocallyConnectedMatrixMultiplyKernel.cpp
index 0da0d4c..6e4c45e 100644
--- a/src/core/CL/kernels/CLLocallyConnectedMatrixMultiplyKernel.cpp
+++ b/src/core/CL/kernels/CLLocallyConnectedMatrixMultiplyKernel.cpp
@@ -23,15 +23,13 @@
  */
 #include "arm_compute/core/CL/kernels/CLLocallyConnectedMatrixMultiplyKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/CL/kernels/CLMagnitudePhaseKernel.cpp b/src/core/CL/kernels/CLMagnitudePhaseKernel.cpp
index ef8ebd5..dc130d0 100644
--- a/src/core/CL/kernels/CLMagnitudePhaseKernel.cpp
+++ b/src/core/CL/kernels/CLMagnitudePhaseKernel.cpp
@@ -27,11 +27,10 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLMaxUnpoolingLayerKernel.cpp b/src/core/CL/kernels/CLMaxUnpoolingLayerKernel.cpp
index 08c7464..a78996d 100644
--- a/src/core/CL/kernels/CLMaxUnpoolingLayerKernel.cpp
+++ b/src/core/CL/kernels/CLMaxUnpoolingLayerKernel.cpp
@@ -23,19 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLMaxUnpoolingLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLMeanStdDevKernel.cpp b/src/core/CL/kernels/CLMeanStdDevKernel.cpp
index 33099c9..5acc3ac 100644
--- a/src/core/CL/kernels/CLMeanStdDevKernel.cpp
+++ b/src/core/CL/kernels/CLMeanStdDevKernel.cpp
@@ -25,14 +25,12 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <cmath>
 #include <set>
diff --git a/src/core/CL/kernels/CLMeanStdDevNormalizationKernel.cpp b/src/core/CL/kernels/CLMeanStdDevNormalizationKernel.cpp
index 5ecbb4b..82a22a9 100644
--- a/src/core/CL/kernels/CLMeanStdDevNormalizationKernel.cpp
+++ b/src/core/CL/kernels/CLMeanStdDevNormalizationKernel.cpp
@@ -25,14 +25,13 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLMedian3x3Kernel.cpp b/src/core/CL/kernels/CLMedian3x3Kernel.cpp
index 5f8c9e5..4b89950 100644
--- a/src/core/CL/kernels/CLMedian3x3Kernel.cpp
+++ b/src/core/CL/kernels/CLMedian3x3Kernel.cpp
@@ -26,8 +26,8 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/CL/kernels/CLMemsetKernel.cpp b/src/core/CL/kernels/CLMemsetKernel.cpp
index f591c2f..186ed2a 100644
--- a/src/core/CL/kernels/CLMemsetKernel.cpp
+++ b/src/core/CL/kernels/CLMemsetKernel.cpp
@@ -23,8 +23,8 @@
  */
 #include "arm_compute/core/CL/kernels/CLMemsetKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLMinMaxLayerKernel.cpp b/src/core/CL/kernels/CLMinMaxLayerKernel.cpp
index 5f0e48d..bf645f8 100644
--- a/src/core/CL/kernels/CLMinMaxLayerKernel.cpp
+++ b/src/core/CL/kernels/CLMinMaxLayerKernel.cpp
@@ -23,14 +23,15 @@
  */
 #include "arm_compute/core/CL/kernels/CLMinMaxLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <climits>
diff --git a/src/core/CL/kernels/CLMinMaxLocationKernel.cpp b/src/core/CL/kernels/CLMinMaxLocationKernel.cpp
index 9bbda40..634b580 100644
--- a/src/core/CL/kernels/CLMinMaxLocationKernel.cpp
+++ b/src/core/CL/kernels/CLMinMaxLocationKernel.cpp
@@ -28,7 +28,8 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <climits>
diff --git a/src/core/CL/kernels/CLNonLinearFilterKernel.cpp b/src/core/CL/kernels/CLNonLinearFilterKernel.cpp
index 16e5113..0a8472b 100644
--- a/src/core/CL/kernels/CLNonLinearFilterKernel.cpp
+++ b/src/core/CL/kernels/CLNonLinearFilterKernel.cpp
@@ -27,12 +27,11 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <cmath>
diff --git a/src/core/CL/kernels/CLNonMaximaSuppression3x3Kernel.cpp b/src/core/CL/kernels/CLNonMaximaSuppression3x3Kernel.cpp
index 958d94c..9c6d44b 100644
--- a/src/core/CL/kernels/CLNonMaximaSuppression3x3Kernel.cpp
+++ b/src/core/CL/kernels/CLNonMaximaSuppression3x3Kernel.cpp
@@ -30,6 +30,7 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <set>
 #include <string>
diff --git a/src/core/CL/kernels/CLNormalizationLayerKernel.cpp b/src/core/CL/kernels/CLNormalizationLayerKernel.cpp
index 7d8e5db..686e6f1 100644
--- a/src/core/CL/kernels/CLNormalizationLayerKernel.cpp
+++ b/src/core/CL/kernels/CLNormalizationLayerKernel.cpp
@@ -23,15 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLNormalizationLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/NormalizationHelpers.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.cpp b/src/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.cpp
index 00bdac3..407ce66 100644
--- a/src/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.cpp
+++ b/src/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.cpp
@@ -23,15 +23,16 @@
  */
 #include "arm_compute/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLPadLayerKernel.cpp b/src/core/CL/kernels/CLPadLayerKernel.cpp
index b2432a0..4572973 100644
--- a/src/core/CL/kernels/CLPadLayerKernel.cpp
+++ b/src/core/CL/kernels/CLPadLayerKernel.cpp
@@ -26,6 +26,8 @@
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLPermuteKernel.cpp b/src/core/CL/kernels/CLPermuteKernel.cpp
index dc2d6fe..6206657 100644
--- a/src/core/CL/kernels/CLPermuteKernel.cpp
+++ b/src/core/CL/kernels/CLPermuteKernel.cpp
@@ -23,8 +23,9 @@
  */
 #include "arm_compute/core/CL/kernels/CLPermuteKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/CL/kernels/CLPixelWiseMultiplicationKernel.cpp b/src/core/CL/kernels/CLPixelWiseMultiplicationKernel.cpp
index 229937e..a7bd4da 100644
--- a/src/core/CL/kernels/CLPixelWiseMultiplicationKernel.cpp
+++ b/src/core/CL/kernels/CLPixelWiseMultiplicationKernel.cpp
@@ -25,11 +25,13 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/Cast.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLPoolingLayerKernel.cpp b/src/core/CL/kernels/CLPoolingLayerKernel.cpp
index 1771834..0570887 100644
--- a/src/core/CL/kernels/CLPoolingLayerKernel.cpp
+++ b/src/core/CL/kernels/CLPoolingLayerKernel.cpp
@@ -23,18 +23,19 @@
  */
 #include "arm_compute/core/CL/kernels/CLPoolingLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLPriorBoxLayerKernel.cpp b/src/core/CL/kernels/CLPriorBoxLayerKernel.cpp
index 3429ef7..202e9fb 100644
--- a/src/core/CL/kernels/CLPriorBoxLayerKernel.cpp
+++ b/src/core/CL/kernels/CLPriorBoxLayerKernel.cpp
@@ -25,13 +25,14 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLQLSTMLayerNormalizationKernel.cpp b/src/core/CL/kernels/CLQLSTMLayerNormalizationKernel.cpp
index 2f676d3..ff6cc86 100644
--- a/src/core/CL/kernels/CLQLSTMLayerNormalizationKernel.cpp
+++ b/src/core/CL/kernels/CLQLSTMLayerNormalizationKernel.cpp
@@ -23,8 +23,9 @@
  */
 #include "arm_compute/core/CL/kernels/CLQLSTMLayerNormalizationKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLQuantizationLayerKernel.cpp b/src/core/CL/kernels/CLQuantizationLayerKernel.cpp
index f6b0888..983cbed 100644
--- a/src/core/CL/kernels/CLQuantizationLayerKernel.cpp
+++ b/src/core/CL/kernels/CLQuantizationLayerKernel.cpp
@@ -23,16 +23,16 @@
  */
 #include "arm_compute/core/CL/kernels/CLQuantizationLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLROIAlignLayerKernel.cpp b/src/core/CL/kernels/CLROIAlignLayerKernel.cpp
index 3f2a904..ca6c6fa 100644
--- a/src/core/CL/kernels/CLROIAlignLayerKernel.cpp
+++ b/src/core/CL/kernels/CLROIAlignLayerKernel.cpp
@@ -23,18 +23,19 @@
  */
 #include "arm_compute/core/CL/kernels/CLROIAlignLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLArray.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute::misc::shape_calculator;
diff --git a/src/core/CL/kernels/CLROIPoolingLayerKernel.cpp b/src/core/CL/kernels/CLROIPoolingLayerKernel.cpp
index c2ed326..55fe5a5 100644
--- a/src/core/CL/kernels/CLROIPoolingLayerKernel.cpp
+++ b/src/core/CL/kernels/CLROIPoolingLayerKernel.cpp
@@ -23,17 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLROIPoolingLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLArray.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cmath>
diff --git a/src/core/CL/kernels/CLRangeKernel.cpp b/src/core/CL/kernels/CLRangeKernel.cpp
index d46cdd7..a4c30b6 100644
--- a/src/core/CL/kernels/CLRangeKernel.cpp
+++ b/src/core/CL/kernels/CLRangeKernel.cpp
@@ -24,9 +24,11 @@
 #include "arm_compute/core/CL/kernels/CLRangeKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Utils.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/CL/kernels/CLReductionOperationKernel.cpp b/src/core/CL/kernels/CLReductionOperationKernel.cpp
index 0ba63cc..325e4b9 100644
--- a/src/core/CL/kernels/CLReductionOperationKernel.cpp
+++ b/src/core/CL/kernels/CLReductionOperationKernel.cpp
@@ -23,17 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLReductionOperationKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLRemapKernel.cpp b/src/core/CL/kernels/CLRemapKernel.cpp
index fe8c81a..8d3f41b 100644
--- a/src/core/CL/kernels/CLRemapKernel.cpp
+++ b/src/core/CL/kernels/CLRemapKernel.cpp
@@ -23,15 +23,14 @@
  */
 #include "arm_compute/core/CL/kernels/CLRemapKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 
diff --git a/src/core/CL/kernels/CLReorgLayerKernel.cpp b/src/core/CL/kernels/CLReorgLayerKernel.cpp
index ab81a8f..ade7761 100644
--- a/src/core/CL/kernels/CLReorgLayerKernel.cpp
+++ b/src/core/CL/kernels/CLReorgLayerKernel.cpp
@@ -26,12 +26,12 @@
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <string>
diff --git a/src/core/CL/kernels/CLReshapeLayerKernel.cpp b/src/core/CL/kernels/CLReshapeLayerKernel.cpp
index 3daf21a..e089709 100644
--- a/src/core/CL/kernels/CLReshapeLayerKernel.cpp
+++ b/src/core/CL/kernels/CLReshapeLayerKernel.cpp
@@ -23,18 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLReshapeLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/Cast.h"
 
 #include <string>
 
diff --git a/src/core/CL/kernels/CLReverseKernel.cpp b/src/core/CL/kernels/CLReverseKernel.cpp
index 6546ced..f824098 100644
--- a/src/core/CL/kernels/CLReverseKernel.cpp
+++ b/src/core/CL/kernels/CLReverseKernel.cpp
@@ -25,12 +25,12 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLScaleKernel.cpp b/src/core/CL/kernels/CLScaleKernel.cpp
index 2e7ee36..8233f21 100644
--- a/src/core/CL/kernels/CLScaleKernel.cpp
+++ b/src/core/CL/kernels/CLScaleKernel.cpp
@@ -23,16 +23,17 @@
  */
 #include "arm_compute/core/CL/kernels/CLScaleKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include "src/core/utils/ScaleUtils.h"
diff --git a/src/core/CL/kernels/CLScharr3x3Kernel.cpp b/src/core/CL/kernels/CLScharr3x3Kernel.cpp
index 3172966..1e33af3 100644
--- a/src/core/CL/kernels/CLScharr3x3Kernel.cpp
+++ b/src/core/CL/kernels/CLScharr3x3Kernel.cpp
@@ -26,11 +26,9 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <set>
 #include <string>
diff --git a/src/core/CL/kernels/CLSelectKernel.cpp b/src/core/CL/kernels/CLSelectKernel.cpp
index dcac78c..d9a1044 100644
--- a/src/core/CL/kernels/CLSelectKernel.cpp
+++ b/src/core/CL/kernels/CLSelectKernel.cpp
@@ -25,12 +25,13 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLSobel3x3Kernel.cpp b/src/core/CL/kernels/CLSobel3x3Kernel.cpp
index 86dcf22..89e5207 100644
--- a/src/core/CL/kernels/CLSobel3x3Kernel.cpp
+++ b/src/core/CL/kernels/CLSobel3x3Kernel.cpp
@@ -26,11 +26,9 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLSobel5x5Kernel.cpp b/src/core/CL/kernels/CLSobel5x5Kernel.cpp
index e010fdd..3e765e4 100644
--- a/src/core/CL/kernels/CLSobel5x5Kernel.cpp
+++ b/src/core/CL/kernels/CLSobel5x5Kernel.cpp
@@ -26,11 +26,9 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLSobel7x7Kernel.cpp b/src/core/CL/kernels/CLSobel7x7Kernel.cpp
index c2b4bec..37ceaba 100644
--- a/src/core/CL/kernels/CLSobel7x7Kernel.cpp
+++ b/src/core/CL/kernels/CLSobel7x7Kernel.cpp
@@ -26,11 +26,9 @@
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLSoftmaxLayerKernel.cpp b/src/core/CL/kernels/CLSoftmaxLayerKernel.cpp
index c7881b9..5c0acda 100644
--- a/src/core/CL/kernels/CLSoftmaxLayerKernel.cpp
+++ b/src/core/CL/kernels/CLSoftmaxLayerKernel.cpp
@@ -23,18 +23,19 @@
  */
 #include "arm_compute/core/CL/kernels/CLSoftmaxLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/KernelDescriptors.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/CL/kernels/CLSpaceToBatchLayerKernel.cpp b/src/core/CL/kernels/CLSpaceToBatchLayerKernel.cpp
index 3e0ac74..c6f70c3 100644
--- a/src/core/CL/kernels/CLSpaceToBatchLayerKernel.cpp
+++ b/src/core/CL/kernels/CLSpaceToBatchLayerKernel.cpp
@@ -24,9 +24,11 @@
 #include "arm_compute/core/CL/kernels/CLSpaceToBatchLayerKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute::misc::shape_calculator;
diff --git a/src/core/CL/kernels/CLSpaceToDepthLayerKernel.cpp b/src/core/CL/kernels/CLSpaceToDepthLayerKernel.cpp
index 4b6c1be..2d46aad 100644
--- a/src/core/CL/kernels/CLSpaceToDepthLayerKernel.cpp
+++ b/src/core/CL/kernels/CLSpaceToDepthLayerKernel.cpp
@@ -24,9 +24,11 @@
 #include "arm_compute/core/CL/kernels/CLSpaceToDepthLayerKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute::misc::shape_calculator;
diff --git a/src/core/CL/kernels/CLStackLayerKernel.cpp b/src/core/CL/kernels/CLStackLayerKernel.cpp
index c283c44..5055065 100644
--- a/src/core/CL/kernels/CLStackLayerKernel.cpp
+++ b/src/core/CL/kernels/CLStackLayerKernel.cpp
@@ -25,13 +25,14 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLStridedSliceKernel.cpp b/src/core/CL/kernels/CLStridedSliceKernel.cpp
index f7b7290..b632e05 100644
--- a/src/core/CL/kernels/CLStridedSliceKernel.cpp
+++ b/src/core/CL/kernels/CLStridedSliceKernel.cpp
@@ -24,10 +24,12 @@
 #include "arm_compute/core/CL/kernels/CLStridedSliceKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/utils/helpers/bit_ops.h"
 #include "arm_compute/core/utils/helpers/tensor_transform.h"
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/utils/helpers/bit_ops.h"
+#include "support/Cast.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLTileKernel.cpp b/src/core/CL/kernels/CLTileKernel.cpp
index bba1525..43c8953 100644
--- a/src/core/CL/kernels/CLTileKernel.cpp
+++ b/src/core/CL/kernels/CLTileKernel.cpp
@@ -23,8 +23,9 @@
  */
 #include "arm_compute/core/CL/kernels/CLTileKernel.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLTransposeKernel.cpp b/src/core/CL/kernels/CLTransposeKernel.cpp
index a47d956..bd91019 100644
--- a/src/core/CL/kernels/CLTransposeKernel.cpp
+++ b/src/core/CL/kernels/CLTransposeKernel.cpp
@@ -23,18 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLTransposeKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/AccessWindowTranspose.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/AccessWindowTranspose.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <set>
 #include <sstream>
diff --git a/src/core/CL/kernels/CLUpsampleLayerKernel.cpp b/src/core/CL/kernels/CLUpsampleLayerKernel.cpp
index 1010550..a4fc10f 100644
--- a/src/core/CL/kernels/CLUpsampleLayerKernel.cpp
+++ b/src/core/CL/kernels/CLUpsampleLayerKernel.cpp
@@ -23,16 +23,18 @@
  */
 #include "arm_compute/core/CL/kernels/CLUpsampleLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLWarpAffineKernel.cpp b/src/core/CL/kernels/CLWarpAffineKernel.cpp
index e8da803..95a7c1b 100644
--- a/src/core/CL/kernels/CLWarpAffineKernel.cpp
+++ b/src/core/CL/kernels/CLWarpAffineKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/CL/kernels/CLWarpAffineKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
@@ -32,6 +31,8 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cstddef>
diff --git a/src/core/CL/kernels/CLWarpPerspectiveKernel.cpp b/src/core/CL/kernels/CLWarpPerspectiveKernel.cpp
index dc7c359..2fe1feb 100644
--- a/src/core/CL/kernels/CLWarpPerspectiveKernel.cpp
+++ b/src/core/CL/kernels/CLWarpPerspectiveKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/CL/kernels/CLWarpPerspectiveKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
 #include "arm_compute/core/CL/ICLTensor.h"
@@ -32,6 +31,8 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <cstddef>
 #include <set>
diff --git a/src/core/CL/kernels/CLWeightsReshapeKernel.cpp b/src/core/CL/kernels/CLWeightsReshapeKernel.cpp
index 267957e..f6996726 100644
--- a/src/core/CL/kernels/CLWeightsReshapeKernel.cpp
+++ b/src/core/CL/kernels/CLWeightsReshapeKernel.cpp
@@ -25,6 +25,8 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CL/kernels/CLWidthConcatenate2TensorsKernel.cpp b/src/core/CL/kernels/CLWidthConcatenate2TensorsKernel.cpp
index 76100c2..27c6508 100644
--- a/src/core/CL/kernels/CLWidthConcatenate2TensorsKernel.cpp
+++ b/src/core/CL/kernels/CLWidthConcatenate2TensorsKernel.cpp
@@ -23,16 +23,16 @@
  */
 #include "arm_compute/core/CL/kernels/CLWidthConcatenate2TensorsKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/helpers/tensor_info.h"
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/utils/helpers/tensor_info.h"
+#include "support/Cast.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLWidthConcatenate4TensorsKernel.cpp b/src/core/CL/kernels/CLWidthConcatenate4TensorsKernel.cpp
index 0377eb7..5ef2cc4 100644
--- a/src/core/CL/kernels/CLWidthConcatenate4TensorsKernel.cpp
+++ b/src/core/CL/kernels/CLWidthConcatenate4TensorsKernel.cpp
@@ -23,17 +23,16 @@
  */
 #include "arm_compute/core/CL/kernels/CLWidthConcatenate4TensorsKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/helpers/tensor_info.h"
-#include "arm_compute/core/utils/misc/Cast.h"
-#include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/utils/helpers/tensor_info.h"
+#include "support/Cast.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLWidthConcatenateLayerKernel.cpp b/src/core/CL/kernels/CLWidthConcatenateLayerKernel.cpp
index d40597f..8b0aaf3 100644
--- a/src/core/CL/kernels/CLWidthConcatenateLayerKernel.cpp
+++ b/src/core/CL/kernels/CLWidthConcatenateLayerKernel.cpp
@@ -25,13 +25,12 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/misc/Cast.h"
-#include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/Cast.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp b/src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp
index 4a1c48a..7fab208 100644
--- a/src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp
+++ b/src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp
@@ -23,10 +23,8 @@
  */
 #include "arm_compute/core/CL/kernels/CLWinogradFilterTransformKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
@@ -36,6 +34,10 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLWinogradInputTransformKernel.cpp b/src/core/CL/kernels/CLWinogradInputTransformKernel.cpp
index c4c2b08..15c239e 100644
--- a/src/core/CL/kernels/CLWinogradInputTransformKernel.cpp
+++ b/src/core/CL/kernels/CLWinogradInputTransformKernel.cpp
@@ -23,10 +23,8 @@
  */
 #include "arm_compute/core/CL/kernels/CLWinogradInputTransformKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/OpenCL.h"
 #include "arm_compute/core/Error.h"
@@ -34,6 +32,10 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/CL/kernels/CLWinogradOutputTransformKernel.cpp b/src/core/CL/kernels/CLWinogradOutputTransformKernel.cpp
index 19f61b1..8ae0255 100644
--- a/src/core/CL/kernels/CLWinogradOutputTransformKernel.cpp
+++ b/src/core/CL/kernels/CLWinogradOutputTransformKernel.cpp
@@ -23,10 +23,8 @@
  */
 #include "arm_compute/core/CL/kernels/CLWinogradOutputTransformKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
@@ -36,6 +34,10 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/CL/kernels/CLYOLOLayerKernel.cpp b/src/core/CL/kernels/CLYOLOLayerKernel.cpp
index 132d5d1..0c7588d 100644
--- a/src/core/CL/kernels/CLYOLOLayerKernel.cpp
+++ b/src/core/CL/kernels/CLYOLOLayerKernel.cpp
@@ -24,17 +24,19 @@
 #include "arm_compute/core/CL/kernels/CLYOLOLayerKernel.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
+#include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/TensorInfo.h"
+#include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
-#include "arm_compute/core/CL/CLHelpers.h"
-#include "arm_compute/core/Types.h"
 #include "support/StringSupport.h"
 
 namespace arm_compute
diff --git a/src/core/CPP/ICPPSimpleKernel.cpp b/src/core/CPP/ICPPSimpleKernel.cpp
index 126bf54..9e4df5e 100644
--- a/src/core/CPP/ICPPSimpleKernel.cpp
+++ b/src/core/CPP/ICPPSimpleKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2018 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -26,6 +26,7 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/ITensor.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
@@ -71,4 +72,4 @@
     return Status{};
 }
 
-} // namespace arm_compute
\ No newline at end of file
+} // namespace arm_compute
diff --git a/arm_compute/core/CPP/Validate.h b/src/core/CPP/Validate.h
similarity index 100%
rename from arm_compute/core/CPP/Validate.h
rename to src/core/CPP/Validate.h
diff --git a/src/core/CPP/kernels/CPPBoxWithNonMaximaSuppressionLimitKernel.cpp b/src/core/CPP/kernels/CPPBoxWithNonMaximaSuppressionLimitKernel.cpp
index 917a6ad..fb17542 100644
--- a/src/core/CPP/kernels/CPPBoxWithNonMaximaSuppressionLimitKernel.cpp
+++ b/src/core/CPP/kernels/CPPBoxWithNonMaximaSuppressionLimitKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,8 +23,8 @@
  */
 #include "arm_compute/core/CPP/kernels/CPPBoxWithNonMaximaSuppressionLimitKernel.h"
 
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <cmath>
diff --git a/src/core/CPP/kernels/CPPCornerCandidatesKernel.cpp b/src/core/CPP/kernels/CPPCornerCandidatesKernel.cpp
index a0cfb3b..d7af9c9 100644
--- a/src/core/CPP/kernels/CPPCornerCandidatesKernel.cpp
+++ b/src/core/CPP/kernels/CPPCornerCandidatesKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,17 +23,9 @@
  */
 #include "arm_compute/core/CPP/kernels/CPPCornerCandidatesKernel.h"
 
-#include "arm_compute/core/Coordinates.h"
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
-#include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
-
-#include "support/Mutex.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute;
 
diff --git a/src/core/CPP/kernels/CPPDetectionWindowNonMaximaSuppressionKernel.cpp b/src/core/CPP/kernels/CPPDetectionWindowNonMaximaSuppressionKernel.cpp
index ec03b72..3166fab 100644
--- a/src/core/CPP/kernels/CPPDetectionWindowNonMaximaSuppressionKernel.cpp
+++ b/src/core/CPP/kernels/CPPDetectionWindowNonMaximaSuppressionKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2018 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/CPP/kernels/CPPDetectionWindowNonMaximaSuppressionKernel.h"
 
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 
 #include <algorithm>
diff --git a/src/core/CPP/kernels/CPPNonMaximumSuppressionKernel.cpp b/src/core/CPP/kernels/CPPNonMaximumSuppressionKernel.cpp
index 89e3058..c1187ff 100644
--- a/src/core/CPP/kernels/CPPNonMaximumSuppressionKernel.cpp
+++ b/src/core/CPP/kernels/CPPNonMaximumSuppressionKernel.cpp
@@ -23,10 +23,12 @@
  */
 #include "arm_compute/core/CPP/kernels/CPPNonMaximumSuppressionKernel.h"
 
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
 
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+
 #include <algorithm>
 
 namespace arm_compute
diff --git a/src/core/CPP/kernels/CPPPermuteKernel.cpp b/src/core/CPP/kernels/CPPPermuteKernel.cpp
index 1d1f0cd..054c7bf 100644
--- a/src/core/CPP/kernels/CPPPermuteKernel.cpp
+++ b/src/core/CPP/kernels/CPPPermuteKernel.cpp
@@ -23,13 +23,10 @@
  */
 #include "arm_compute/core/CPP/kernels/CPPPermuteKernel.h"
 
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/ITensor.h"
-#include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
-#include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <cstddef>
 #include <cstdint>
diff --git a/src/core/CPP/kernels/CPPTopKVKernel.cpp b/src/core/CPP/kernels/CPPTopKVKernel.cpp
index 7ba8d7c..d2b54e4 100644
--- a/src/core/CPP/kernels/CPPTopKVKernel.cpp
+++ b/src/core/CPP/kernels/CPPTopKVKernel.cpp
@@ -22,16 +22,14 @@
  * SOFTWARE.
  */
 #include "arm_compute/core/CPP/kernels/CPPTopKVKernel.h"
-#include "arm_compute/core/Coordinates.h"
-#include "arm_compute/core/Error.h"
+
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
-#include "arm_compute/core/Utils.h"
-#include "arm_compute/core/Validate.h"
-#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/Traits.h"
 
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+
 namespace arm_compute
 {
 namespace
diff --git a/src/core/CPP/kernels/CPPUpsampleKernel.cpp b/src/core/CPP/kernels/CPPUpsampleKernel.cpp
index ff4ffb6..7ef83fb 100644
--- a/src/core/CPP/kernels/CPPUpsampleKernel.cpp
+++ b/src/core/CPP/kernels/CPPUpsampleKernel.cpp
@@ -23,13 +23,8 @@
  */
 #include "arm_compute/core/CPP/kernels/CPPUpsampleKernel.h"
 
-#include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
-#include "arm_compute/core/ITensor.h"
-#include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Types.h"
-#include "arm_compute/core/Validate.h"
-#include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <cstddef>
 #include <cstdint>
diff --git a/src/core/GLES_COMPUTE/IGCSimpleKernel.cpp b/src/core/GLES_COMPUTE/IGCSimpleKernel.cpp
index 6609f45..fb31ac8 100644
--- a/src/core/GLES_COMPUTE/IGCSimpleKernel.cpp
+++ b/src/core/GLES_COMPUTE/IGCSimpleKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -27,6 +27,7 @@
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute;
 
diff --git a/src/core/GLES_COMPUTE/kernels/GCAbsoluteDifferenceKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCAbsoluteDifferenceKernel.cpp
index f0a5003..5e8accc 100644
--- a/src/core/GLES_COMPUTE/kernels/GCAbsoluteDifferenceKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCAbsoluteDifferenceKernel.cpp
@@ -32,6 +32,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/GLES_COMPUTE/kernels/GCActivationLayerKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCActivationLayerKernel.cpp
index 1c02f41..0173b81 100644
--- a/src/core/GLES_COMPUTE/kernels/GCActivationLayerKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCActivationLayerKernel.cpp
@@ -32,6 +32,8 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/GLES_COMPUTE/kernels/GCArithmeticAdditionKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCArithmeticAdditionKernel.cpp
index 06c3486..f31c8ca 100644
--- a/src/core/GLES_COMPUTE/kernels/GCArithmeticAdditionKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCArithmeticAdditionKernel.cpp
@@ -34,6 +34,8 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cstddef>
diff --git a/src/core/GLES_COMPUTE/kernels/GCBatchNormalizationLayerKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCBatchNormalizationLayerKernel.cpp
index 3bd34ac..9281ce5 100644
--- a/src/core/GLES_COMPUTE/kernels/GCBatchNormalizationLayerKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCBatchNormalizationLayerKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCBatchNormalizationLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
 #include "arm_compute/core/GLES_COMPUTE/IGCTensor.h"
@@ -31,6 +30,9 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/GLES_COMPUTE/kernels/GCCol2ImKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCCol2ImKernel.cpp
index 4fe6484..5781c56 100644
--- a/src/core/GLES_COMPUTE/kernels/GCCol2ImKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCCol2ImKernel.cpp
@@ -24,7 +24,6 @@
 
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCCol2ImKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
@@ -33,6 +32,9 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
@@ -42,7 +44,7 @@
 {
 }
 
-void GCCol2ImKernel::configure(const IGCTensor *input, IGCTensor *output,
+void GCCol2ImKernel::configure(const IGCTensor *input, IGCTensor    *output,
                                std::pair<unsigned int, unsigned int> convolved_dims)
 {
     ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::F16, DataType::F32);
diff --git a/src/core/GLES_COMPUTE/kernels/GCDepthConcatenateLayerKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCDepthConcatenateLayerKernel.cpp
index 458cb63..3256f11 100644
--- a/src/core/GLES_COMPUTE/kernels/GCDepthConcatenateLayerKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCDepthConcatenateLayerKernel.cpp
@@ -32,6 +32,8 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/GLES_COMPUTE/kernels/GCDepthwiseConvolutionLayer3x3Kernel.cpp b/src/core/GLES_COMPUTE/kernels/GCDepthwiseConvolutionLayer3x3Kernel.cpp
index cb70dae..95d487b 100644
--- a/src/core/GLES_COMPUTE/kernels/GCDepthwiseConvolutionLayer3x3Kernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCDepthwiseConvolutionLayer3x3Kernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCDepthwiseConvolutionLayer3x3Kernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
@@ -34,6 +33,9 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.cpp
index 302b21b..9ce8ace 100644
--- a/src/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
@@ -33,6 +32,9 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
@@ -44,7 +46,7 @@
 }
 
 template <unsigned int kernel_size>
-BorderSize GCDirectConvolutionLayerKernel<kernel_size>::border_size() const
+BorderSize             GCDirectConvolutionLayerKernel<kernel_size>::border_size() const
 {
     return _border_size;
 }
@@ -70,8 +72,8 @@
     }
 
     // Get convolved dimensions
-    unsigned int owidth       = 0;
-    unsigned int oheight      = 0;
+    unsigned int owidth  = 0;
+    unsigned int oheight = 0;
     std::tie(owidth, oheight) = scaled_dimensions(input->info()->dimension(0), input->info()->dimension(1), kernel_size, kernel_size, conv_info);
 
     TensorShape output_shape = input->info()->tensor_shape();
@@ -238,20 +240,20 @@
                 num_elems_written_per_iteration_x = 4;
 #elif defined(PROCESS_4X_2Y_1Z)
                 options.emplace("#define PROCESS_4X_2Y_1Z");
-                num_elems_read_per_iteration_x = 4;
-                num_elems_read_per_iteration_y = 2;
+                num_elems_read_per_iteration_x    = 4;
+                num_elems_read_per_iteration_y    = 2;
                 num_elems_written_per_iteration_x = 4;
                 num_elems_written_per_iteration_y = 2;
 #elif defined(PROCESS_4X_3Y_1Z)
                 options.emplace("#define PROCESS_4X_3Y_1Z");
-                num_elems_read_per_iteration_x = 4;
-                num_elems_read_per_iteration_y = 3;
+                num_elems_read_per_iteration_x    = 4;
+                num_elems_read_per_iteration_y    = 3;
                 num_elems_written_per_iteration_x = 4;
                 num_elems_written_per_iteration_y = 3;
 #elif defined(PROCESS_4X_4Y_1Z)
                 options.emplace("#define PROCESS_4X_4Y_1Z");
-                num_elems_read_per_iteration_x = 4;
-                num_elems_read_per_iteration_y = 4;
+                num_elems_read_per_iteration_x    = 4;
+                num_elems_read_per_iteration_y    = 4;
                 num_elems_written_per_iteration_x = 4;
                 num_elems_written_per_iteration_y = 4;
 #elif defined(PROCESS_4X_2Y_2Z)
diff --git a/src/core/GLES_COMPUTE/kernels/GCDropoutLayerKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCDropoutLayerKernel.cpp
index 5c6722a..bda6599 100644
--- a/src/core/GLES_COMPUTE/kernels/GCDropoutLayerKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCDropoutLayerKernel.cpp
@@ -32,6 +32,8 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cmath>
diff --git a/src/core/GLES_COMPUTE/kernels/GCFillBorderKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCFillBorderKernel.cpp
index 3b3118b..7ffcdd2 100644
--- a/src/core/GLES_COMPUTE/kernels/GCFillBorderKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCFillBorderKernel.cpp
@@ -32,6 +32,8 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cstdint>
diff --git a/src/core/GLES_COMPUTE/kernels/GCGEMMInterleave4x4Kernel.cpp b/src/core/GLES_COMPUTE/kernels/GCGEMMInterleave4x4Kernel.cpp
index e0f7e95..d395759 100644
--- a/src/core/GLES_COMPUTE/kernels/GCGEMMInterleave4x4Kernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCGEMMInterleave4x4Kernel.cpp
@@ -33,6 +33,8 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixAccumulateBiasesKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixAccumulateBiasesKernel.cpp
index c9eb433..66fdde5 100644
--- a/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixAccumulateBiasesKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixAccumulateBiasesKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCGEMMMatrixAccumulateBiasesKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
@@ -33,6 +32,9 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixAdditionKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixAdditionKernel.cpp
index e8298bc..daad70b 100644
--- a/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixAdditionKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixAdditionKernel.cpp
@@ -32,6 +32,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixMultiplyKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixMultiplyKernel.cpp
index dd03faf..2f69728 100644
--- a/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixMultiplyKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCGEMMMatrixMultiplyKernel.cpp
@@ -23,8 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCGEMMMatrixMultiplyKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/AccessWindowTranspose.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
@@ -37,6 +35,10 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/AccessWindowTranspose.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/GLES_COMPUTE/kernels/GCGEMMTranspose1xWKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCGEMMTranspose1xWKernel.cpp
index 4190163..1d6ef3d 100644
--- a/src/core/GLES_COMPUTE/kernels/GCGEMMTranspose1xWKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCGEMMTranspose1xWKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCGEMMTranspose1xWKernel.h"
 
-#include "arm_compute/core/AccessWindowTranspose.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
@@ -33,6 +32,9 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowTranspose.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cmath>
diff --git a/src/core/GLES_COMPUTE/kernels/GCIm2ColKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCIm2ColKernel.cpp
index 64f2d63..c12dd38 100644
--- a/src/core/GLES_COMPUTE/kernels/GCIm2ColKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCIm2ColKernel.cpp
@@ -24,7 +24,6 @@
 
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCIm2ColKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
@@ -35,6 +34,9 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cmath>
@@ -91,7 +93,7 @@
     int stride_y = 0;
 
     std::tie(stride_x, stride_y) = conv_info.stride();
-    _kernel_dims                 = std::make_pair(kernel_dims.width, kernel_dims.height);
+    _kernel_dims = std::make_pair(kernel_dims.width, kernel_dims.height);
 
     const bool run_img2col_reduced = (output->info()->dimension(0) == (input->info()->dimension(0) * input->info()->dimension(1) * input->info()->dimension(2))) && (TensorShape::num_max_dimensions >= 4)
                                      && (std::equal(input->info()->tensor_shape().cbegin() + 3,
@@ -109,9 +111,9 @@
         }
 
         build_opts.emplace("#define IM2COL_GENERIC");
-        _convolved_dims                    = scaled_dimensions(input->info()->dimension(0), input->info()->dimension(1),
-                                                               kernel_dims.width, kernel_dims.height,
-                                                               conv_info, dilation);
+        _convolved_dims = scaled_dimensions(input->info()->dimension(0), input->info()->dimension(1),
+                                            kernel_dims.width, kernel_dims.height,
+                                            conv_info, dilation);
         _num_elems_processed_per_iteration = (input->info()->data_type() == DataType::F32) ? 1 : 2;
 
         build_opts.emplace("#define KERNEL_WIDTH " + support::cpp11::to_string(kernel_dims.width));
diff --git a/src/core/GLES_COMPUTE/kernels/GCNormalizationLayerKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCNormalizationLayerKernel.cpp
index 5fa1987..c29d9fc 100644
--- a/src/core/GLES_COMPUTE/kernels/GCNormalizationLayerKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCNormalizationLayerKernel.cpp
@@ -31,6 +31,8 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <string>
diff --git a/src/core/GLES_COMPUTE/kernels/GCNormalizePlanarYUVLayerKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCNormalizePlanarYUVLayerKernel.cpp
index 6a79990..971b540 100644
--- a/src/core/GLES_COMPUTE/kernels/GCNormalizePlanarYUVLayerKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCNormalizePlanarYUVLayerKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCNormalizePlanarYUVLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
 #include "arm_compute/core/GLES_COMPUTE/IGCTensor.h"
@@ -31,6 +30,9 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/core/GLES_COMPUTE/kernels/GCPixelWiseMultiplicationKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCPixelWiseMultiplicationKernel.cpp
index 45aa06c..7655914 100644
--- a/src/core/GLES_COMPUTE/kernels/GCPixelWiseMultiplicationKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCPixelWiseMultiplicationKernel.cpp
@@ -32,6 +32,8 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <cmath>
diff --git a/src/core/GLES_COMPUTE/kernels/GCPoolingLayerKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCPoolingLayerKernel.cpp
index a592c09..13efd10 100644
--- a/src/core/GLES_COMPUTE/kernels/GCPoolingLayerKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCPoolingLayerKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCPoolingLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
 #include "arm_compute/core/GLES_COMPUTE/IGCTensor.h"
@@ -33,6 +32,9 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/GLES_COMPUTE/kernels/GCScaleKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCScaleKernel.cpp
index cf10b92..a0795c6 100644
--- a/src/core/GLES_COMPUTE/kernels/GCScaleKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCScaleKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCScaleKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
@@ -33,6 +32,9 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/GLES_COMPUTE/kernels/GCSoftmaxLayerKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCSoftmaxLayerKernel.cpp
index f4ed961..39d586d 100644
--- a/src/core/GLES_COMPUTE/kernels/GCSoftmaxLayerKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCSoftmaxLayerKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCSoftmaxLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
 #include "arm_compute/core/GLES_COMPUTE/IGCTensor.h"
@@ -33,6 +32,9 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/GLES_COMPUTE/kernels/GCTensorShiftKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCTensorShiftKernel.cpp
index d06be9b..78b0084 100644
--- a/src/core/GLES_COMPUTE/kernels/GCTensorShiftKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCTensorShiftKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCTensorShiftKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
@@ -33,6 +32,9 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/GLES_COMPUTE/kernels/GCTransposeKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCTransposeKernel.cpp
index 66b4a55..3bec05b 100644
--- a/src/core/GLES_COMPUTE/kernels/GCTransposeKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCTransposeKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/GLES_COMPUTE/kernels/GCTransposeKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
 #include "arm_compute/core/GLES_COMPUTE/GCKernelLibrary.h"
@@ -31,6 +30,9 @@
 #include "arm_compute/core/GLES_COMPUTE/OpenGLES.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Types.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include <set>
diff --git a/src/core/GLES_COMPUTE/kernels/GCWeightsReshapeKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCWeightsReshapeKernel.cpp
index 9a430b4..bcdbfb6 100644
--- a/src/core/GLES_COMPUTE/kernels/GCWeightsReshapeKernel.cpp
+++ b/src/core/GLES_COMPUTE/kernels/GCWeightsReshapeKernel.cpp
@@ -32,6 +32,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/StringSupport.h"
 
 #include "arm_compute/core/GLES_COMPUTE/GCHelpers.h"
diff --git a/src/core/Helpers.cpp b/src/core/Helpers.cpp
index 6701ff4..e692cc1 100644
--- a/src/core/Helpers.cpp
+++ b/src/core/Helpers.cpp
@@ -25,162 +25,6 @@
 
 namespace arm_compute
 {
-Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
-{
-    if(!skip_border)
-    {
-        border_size = BorderSize(0);
-    }
-
-    const Coordinates &anchor = valid_region.anchor;
-    const TensorShape &shape  = valid_region.shape;
-
-    Window window;
-
-    window.set(0, Window::Dimension(
-                   // Skip the border left of the image
-                   anchor[0] + border_size.left,
-                   // Skip the border right of the image
-                   // Make sure the window width is a multiple of the step size
-                   anchor[0] + border_size.left + ceil_to_multiple(std::max(0, static_cast<int>(shape[0]) - static_cast<int>(border_size.left) - static_cast<int>(border_size.right)), steps[0]),
-                   steps[0]));
-
-    size_t n = 1;
-
-    if(anchor.num_dimensions() > 1)
-    {
-        window.set(1, Window::Dimension(
-                       // Skip the border above the image
-                       anchor[1] + border_size.top,
-                       // Skip the border below the image
-                       anchor[1] + border_size.top + ceil_to_multiple(std::max(0, static_cast<int>(shape[1]) - static_cast<int>(border_size.top) - static_cast<int>(border_size.bottom)), steps[1]),
-                       steps[1]));
-
-        ++n;
-    }
-
-    if(anchor.num_dimensions() > 2)
-    {
-        window.set(2, Window::Dimension(anchor[2], std::max<size_t>(1, shape[2]), steps[2]));
-
-        ++n;
-    }
-
-    for(; n < anchor.num_dimensions(); ++n)
-    {
-        window.set(n, Window::Dimension(anchor[n], std::max<size_t>(1, shape[n])));
-    }
-
-    for(; n < Coordinates::num_max_dimensions; ++n)
-    {
-        window.set(n, Window::Dimension(0, 1));
-    }
-
-    return window;
-}
-
-Window calculate_max_enlarged_window(const ValidRegion &valid_region, const Steps &steps, BorderSize border_size)
-{
-    const Coordinates &anchor = valid_region.anchor;
-    const TensorShape &shape  = valid_region.shape;
-
-    Window window;
-
-    window.set(0, Window::Dimension(
-                   // move the anchor to the start from the border
-                   anchor[0] - border_size.left,
-                   // move the anchor to include the right end border
-                   // Make sure the window width is a multiple of the step size
-                   anchor[0] - border_size.left + ceil_to_multiple(shape[0] + border_size.left + border_size.right, steps[0]),
-                   steps[0]));
-
-    size_t n = 1;
-
-    if(anchor.num_dimensions() > 1)
-    {
-        window.set(1, Window::Dimension(
-                       // Include the border above the image
-                       anchor[1] - border_size.top,
-                       // Include the border below the image
-                       anchor[1] - border_size.top + ceil_to_multiple(shape[1] + border_size.top + border_size.bottom, steps[1]),
-                       steps[1]));
-
-        ++n;
-    }
-
-    if(anchor.num_dimensions() > 2)
-    {
-        window.set(2, Window::Dimension(0, std::max<size_t>(1, shape[n]), steps[2]));
-
-        ++n;
-    }
-
-    for(; n < anchor.num_dimensions(); ++n)
-    {
-        window.set(n, Window::Dimension(anchor[n], std::max<size_t>(1, shape[n])));
-    }
-
-    for(; n < Coordinates::num_max_dimensions; ++n)
-    {
-        window.set(n, Window::Dimension(0, 1));
-    }
-
-    return window;
-}
-
-Window calculate_max_window_horizontal(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
-{
-    if(skip_border)
-    {
-        border_size.top    = 0;
-        border_size.bottom = 0;
-    }
-    else
-    {
-        border_size.left  = 0;
-        border_size.right = 0;
-    }
-
-    const Coordinates &anchor = valid_region.anchor;
-    const TensorShape &shape  = valid_region.shape;
-
-    Window window;
-
-    window.set(0, Window::Dimension(
-                   // Skip the border left of the image
-                   anchor[0] + border_size.left,
-                   // Skip the border right of the image
-                   // Make sure the window width is a multiple of the step size
-                   anchor[0] + border_size.left + ceil_to_multiple(std::max(0, static_cast<int>(shape[0]) - static_cast<int>(border_size.left) - static_cast<int>(border_size.right)), steps[0]),
-                   steps[0]));
-
-    size_t n = 1;
-
-    if(anchor.num_dimensions() > 1)
-    {
-        window.set(1, Window::Dimension(
-                       // Skip the border above the image
-                       anchor[1] - border_size.top,
-                       // Skip the border below the image
-                       anchor[1] + shape[1] + border_size.bottom,
-                       1));
-
-        ++n;
-    }
-
-    for(; n < anchor.num_dimensions(); ++n)
-    {
-        window.set(n, Window::Dimension(anchor[n], std::max<size_t>(1, shape[n])));
-    }
-
-    for(; n < Coordinates::num_max_dimensions; ++n)
-    {
-        window.set(n, Window::Dimension(0, 1));
-    }
-
-    return window;
-}
-
 ValidRegion calculate_valid_region_scale(const ITensorInfo &src_info, const TensorShape &dst_shape,
                                          InterpolationPolicy interpolate_policy, SamplingPolicy sampling_policy, bool border_undefined)
 {
@@ -256,19 +100,4 @@
 
     return valid_region;
 }
-
-PermutationVector get_permutation_vector_from_softmax_axis(size_t actual_axis)
-{
-    switch(actual_axis)
-    {
-        case 1:
-            return PermutationVector(1U, 0U, 2U, 3U);
-        case 2:
-            return PermutationVector(2U, 1U, 0U, 3U);
-        case 3:
-            return PermutationVector(3U, 1U, 2U, 0U);
-        default:
-            ARM_COMPUTE_ERROR("Axis not supported");
-    }
-}
 } // namespace arm_compute
\ No newline at end of file
diff --git a/src/core/NEON/NETracePoint.cpp b/src/core/NEON/NETracePoint.cpp
index cb0dc14..4a6bffa 100644
--- a/src/core/NEON/NETracePoint.cpp
+++ b/src/core/NEON/NETracePoint.cpp
@@ -24,8 +24,8 @@
 #include "arm_compute/core/TracePoint.h"
 
 #include "arm_compute/core/NEON/kernels/NELKTrackerKernel.h"
-#include "arm_compute/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h"
-#include "arm_compute/core/NEON/kernels/convolution/common/convolution.hpp"
+#include "src/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h"
+#include "src/core/NEON/kernels/convolution/common/convolution.hpp"
 #include "utils/TypePrinter.h"
 
 #include <memory>
diff --git a/src/core/NEON/kernels/NEAbsoluteDifferenceKernel.cpp b/src/core/NEON/kernels/NEAbsoluteDifferenceKernel.cpp
index 3d4800f..acea0af 100644
--- a/src/core/NEON/kernels/NEAbsoluteDifferenceKernel.cpp
+++ b/src/core/NEON/kernels/NEAbsoluteDifferenceKernel.cpp
@@ -30,6 +30,8 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEAccumulateKernel.cpp b/src/core/NEON/kernels/NEAccumulateKernel.cpp
index 7c85f69..73ef7eb 100644
--- a/src/core/NEON/kernels/NEAccumulateKernel.cpp
+++ b/src/core/NEON/kernels/NEAccumulateKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -28,6 +28,8 @@
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEActivationLayerKernel.cpp b/src/core/NEON/kernels/NEActivationLayerKernel.cpp
index d80aab7..9616f4f 100644
--- a/src/core/NEON/kernels/NEActivationLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEActivationLayerKernel.cpp
@@ -23,15 +23,17 @@
  */
 #include "arm_compute/core/NEON/kernels/NEActivationLayerKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NESymm.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <set>
diff --git a/src/core/NEON/kernels/NEArithmeticAdditionKernel.cpp b/src/core/NEON/kernels/NEArithmeticAdditionKernel.cpp
index 525e286..7f1a35f 100644
--- a/src/core/NEON/kernels/NEArithmeticAdditionKernel.cpp
+++ b/src/core/NEON/kernels/NEArithmeticAdditionKernel.cpp
@@ -23,12 +23,14 @@
  */
 #include "arm_compute/core/NEON/kernels/NEArithmeticAdditionKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <map>
 #include <string>
diff --git a/src/core/NEON/kernels/NEArithmeticSubtractionKernel.cpp b/src/core/NEON/kernels/NEArithmeticSubtractionKernel.cpp
index a3da750..49e503f 100644
--- a/src/core/NEON/kernels/NEArithmeticSubtractionKernel.cpp
+++ b/src/core/NEON/kernels/NEArithmeticSubtractionKernel.cpp
@@ -23,12 +23,14 @@
  */
 #include "arm_compute/core/NEON/kernels/NEArithmeticSubtractionKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NESymm.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEBatchConcatenateLayerKernel.cpp b/src/core/NEON/kernels/NEBatchConcatenateLayerKernel.cpp
index c7169d8..65ac996 100644
--- a/src/core/NEON/kernels/NEBatchConcatenateLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEBatchConcatenateLayerKernel.cpp
@@ -32,6 +32,8 @@
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEBatchNormalizationLayerKernel.cpp b/src/core/NEON/kernels/NEBatchNormalizationLayerKernel.cpp
index 50e4647..bda3966 100644
--- a/src/core/NEON/kernels/NEBatchNormalizationLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEBatchNormalizationLayerKernel.cpp
@@ -23,14 +23,16 @@
  */
 #include "arm_compute/core/NEON/kernels/NEBatchNormalizationLayerKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/NEMath.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "src/core/NEON/kernels/detail/NEActivationFunctionDetail.h"
 #include "src/core/NEON/wrapper/wrapper.h"
diff --git a/src/core/NEON/kernels/NEBatchToSpaceLayerKernel.cpp b/src/core/NEON/kernels/NEBatchToSpaceLayerKernel.cpp
index eb28ce0..e24d7b6 100644
--- a/src/core/NEON/kernels/NEBatchToSpaceLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEBatchToSpaceLayerKernel.cpp
@@ -28,6 +28,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute::misc::shape_calculator;
 
diff --git a/src/core/NEON/kernels/NEBitwiseAndKernel.cpp b/src/core/NEON/kernels/NEBitwiseAndKernel.cpp
index caaa6c2..2d49ff8 100644
--- a/src/core/NEON/kernels/NEBitwiseAndKernel.cpp
+++ b/src/core/NEON/kernels/NEBitwiseAndKernel.cpp
@@ -28,6 +28,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstdint>
diff --git a/src/core/NEON/kernels/NEBitwiseNotKernel.cpp b/src/core/NEON/kernels/NEBitwiseNotKernel.cpp
index 4da07f9..eed9b27 100644
--- a/src/core/NEON/kernels/NEBitwiseNotKernel.cpp
+++ b/src/core/NEON/kernels/NEBitwiseNotKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -27,6 +27,8 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstdint>
diff --git a/src/core/NEON/kernels/NEBitwiseOrKernel.cpp b/src/core/NEON/kernels/NEBitwiseOrKernel.cpp
index 591acf5..f96117e 100644
--- a/src/core/NEON/kernels/NEBitwiseOrKernel.cpp
+++ b/src/core/NEON/kernels/NEBitwiseOrKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -27,6 +27,8 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstdint>
diff --git a/src/core/NEON/kernels/NEBitwiseXorKernel.cpp b/src/core/NEON/kernels/NEBitwiseXorKernel.cpp
index b0aec40..45d2b0a 100644
--- a/src/core/NEON/kernels/NEBitwiseXorKernel.cpp
+++ b/src/core/NEON/kernels/NEBitwiseXorKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -27,6 +27,8 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstdint>
diff --git a/src/core/NEON/kernels/NEBoundingBoxTransformKernel.cpp b/src/core/NEON/kernels/NEBoundingBoxTransformKernel.cpp
index 56444dc..5a18e88 100644
--- a/src/core/NEON/kernels/NEBoundingBoxTransformKernel.cpp
+++ b/src/core/NEON/kernels/NEBoundingBoxTransformKernel.cpp
@@ -23,12 +23,14 @@
  */
 #include "arm_compute/core/NEON/kernels/NEBoundingBoxTransformKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEBox3x3Kernel.cpp b/src/core/NEON/kernels/NEBox3x3Kernel.cpp
index d5d03a9..1177f6f 100644
--- a/src/core/NEON/kernels/NEBox3x3Kernel.cpp
+++ b/src/core/NEON/kernels/NEBox3x3Kernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,6 +29,9 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/NEON/INEKernel.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+
 #include <arm_neon.h>
 
 using namespace arm_compute;
diff --git a/src/core/NEON/kernels/NECannyEdgeKernel.cpp b/src/core/NEON/kernels/NECannyEdgeKernel.cpp
index 0278bb0..da33c1b 100644
--- a/src/core/NEON/kernels/NECannyEdgeKernel.cpp
+++ b/src/core/NEON/kernels/NECannyEdgeKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NECannyEdgeKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -31,6 +30,11 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEChannelCombineKernel.cpp b/src/core/NEON/kernels/NEChannelCombineKernel.cpp
index 0de6c43..7bd3808 100644
--- a/src/core/NEON/kernels/NEChannelCombineKernel.cpp
+++ b/src/core/NEON/kernels/NEChannelCombineKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -33,6 +33,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEChannelExtractKernel.cpp b/src/core/NEON/kernels/NEChannelExtractKernel.cpp
index 800c636..86245ac 100644
--- a/src/core/NEON/kernels/NEChannelExtractKernel.cpp
+++ b/src/core/NEON/kernels/NEChannelExtractKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -34,6 +34,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEChannelShuffleLayerKernel.cpp b/src/core/NEON/kernels/NEChannelShuffleLayerKernel.cpp
index 88cd0ae..6d04d71 100644
--- a/src/core/NEON/kernels/NEChannelShuffleLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEChannelShuffleLayerKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEChannelShuffleLayerKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -31,6 +30,9 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NECol2ImKernel.cpp b/src/core/NEON/kernels/NECol2ImKernel.cpp
index 6a07def..f319237 100644
--- a/src/core/NEON/kernels/NECol2ImKernel.cpp
+++ b/src/core/NEON/kernels/NECol2ImKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEColorConvertKernel.cpp b/src/core/NEON/kernels/NEColorConvertKernel.cpp
index bc8c775..f933a2a 100644
--- a/src/core/NEON/kernels/NEColorConvertKernel.cpp
+++ b/src/core/NEON/kernels/NEColorConvertKernel.cpp
@@ -33,6 +33,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "src/core/NEON/kernels/detail/NEColorConvertHelper.inl"
 
diff --git a/src/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.cpp b/src/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.cpp
index 97bb8cc..8716cfd 100644
--- a/src/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.cpp
+++ b/src/core/NEON/kernels/NEConvertFullyConnectedWeightsKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -25,6 +25,8 @@
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Types.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEConvertQuantizedSignednessKernel.cpp b/src/core/NEON/kernels/NEConvertQuantizedSignednessKernel.cpp
index f40f121..bd8ea30 100644
--- a/src/core/NEON/kernels/NEConvertQuantizedSignednessKernel.cpp
+++ b/src/core/NEON/kernels/NEConvertQuantizedSignednessKernel.cpp
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEConvolutionKernel.cpp b/src/core/NEON/kernels/NEConvolutionKernel.cpp
index 7103fa1..69b65b2 100644
--- a/src/core/NEON/kernels/NEConvolutionKernel.cpp
+++ b/src/core/NEON/kernels/NEConvolutionKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -32,6 +32,8 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NECopyKernel.cpp b/src/core/NEON/kernels/NECopyKernel.cpp
index 3d00139..b299957 100644
--- a/src/core/NEON/kernels/NECopyKernel.cpp
+++ b/src/core/NEON/kernels/NECopyKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NECropKernel.cpp b/src/core/NEON/kernels/NECropKernel.cpp
index 7c65e71..5fb55d9 100644
--- a/src/core/NEON/kernels/NECropKernel.cpp
+++ b/src/core/NEON/kernels/NECropKernel.cpp
@@ -23,17 +23,18 @@
  */
 #include "arm_compute/core/NEON/kernels/NECropKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Window.h"
-
 #include "arm_compute/core/Types.h"
-#include "arm_compute/core/utils/helpers/bit_ops.h"
+#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/helpers/tensor_transform.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/utils/helpers/bit_ops.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NECumulativeDistributionKernel.cpp b/src/core/NEON/kernels/NECumulativeDistributionKernel.cpp
index cec0e1c..5628802 100644
--- a/src/core/NEON/kernels/NECumulativeDistributionKernel.cpp
+++ b/src/core/NEON/kernels/NECumulativeDistributionKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2018 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <cmath>
diff --git a/src/core/NEON/kernels/NEDepthConcatenateLayerKernel.cpp b/src/core/NEON/kernels/NEDepthConcatenateLayerKernel.cpp
index 6066326..b500268 100644
--- a/src/core/NEON/kernels/NEDepthConcatenateLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEDepthConcatenateLayerKernel.cpp
@@ -34,6 +34,8 @@
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <cstdint>
 
diff --git a/src/core/NEON/kernels/NEDepthConvertLayerKernel.cpp b/src/core/NEON/kernels/NEDepthConvertLayerKernel.cpp
index ee23909..259ece7 100644
--- a/src/core/NEON/kernels/NEDepthConvertLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEDepthConvertLayerKernel.cpp
@@ -23,17 +23,18 @@
  */
 #include "arm_compute/core/NEON/kernels/NEDepthConvertLayerKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/utils/misc/SaturateCast.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/NEMath.h"
-
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/SaturateCast.h"
 
 using namespace arm_compute;
 
diff --git a/src/core/NEON/kernels/NEDepthToSpaceLayerKernel.cpp b/src/core/NEON/kernels/NEDepthToSpaceLayerKernel.cpp
index 6465848..403e7aa 100644
--- a/src/core/NEON/kernels/NEDepthToSpaceLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEDepthToSpaceLayerKernel.cpp
@@ -29,6 +29,9 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+
 #include <arm_neon.h>
 #include <cstdint>
 
diff --git a/src/core/NEON/kernels/NEDepthwiseConvolutionLayerNativeKernel.cpp b/src/core/NEON/kernels/NEDepthwiseConvolutionLayerNativeKernel.cpp
index 0a34ee6..533b374 100644
--- a/src/core/NEON/kernels/NEDepthwiseConvolutionLayerNativeKernel.cpp
+++ b/src/core/NEON/kernels/NEDepthwiseConvolutionLayerNativeKernel.cpp
@@ -23,13 +23,15 @@
  */
 #include "arm_compute/core/NEON/kernels/NEDepthwiseConvolutionLayerNativeKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/kernels/convolution/depthwise/impl_qa8_qa8.hpp"
 #include "src/core/NEON/wrapper/traits.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/ToolchainSupport.h"
 
 namespace arm_compute
@@ -48,7 +50,6 @@
 
 struct DepthwiseConvolutionRunInfo
 {
-public:
     const size_t   num_read_elements_per_iteration;
     const uint32_t x_start;
     const uint32_t x_end;
diff --git a/src/core/NEON/kernels/NEDequantizationLayerKernel.cpp b/src/core/NEON/kernels/NEDequantizationLayerKernel.cpp
index 9352088..2f3c6f4 100644
--- a/src/core/NEON/kernels/NEDequantizationLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEDequantizationLayerKernel.cpp
@@ -23,16 +23,18 @@
  */
 #include "arm_compute/core/NEON/kernels/NEDequantizationLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NESymm.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEDerivativeKernel.cpp b/src/core/NEON/kernels/NEDerivativeKernel.cpp
index ad590e9..5d3fc01 100644
--- a/src/core/NEON/kernels/NEDerivativeKernel.cpp
+++ b/src/core/NEON/kernels/NEDerivativeKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,6 +29,8 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEDilateKernel.cpp b/src/core/NEON/kernels/NEDilateKernel.cpp
index c30dab2..cc781c6 100644
--- a/src/core/NEON/kernels/NEDilateKernel.cpp
+++ b/src/core/NEON/kernels/NEDilateKernel.cpp
@@ -28,6 +28,8 @@
 #include "arm_compute/core/NEON/INEKernel.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp b/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp
index c22fa6a..56cd6e6 100644
--- a/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp
@@ -26,8 +26,6 @@
 #include "src/core/NEON/kernels/detail/NEDirectConvolutionDetail.h"
 #include "src/core/NEON/wrapper/wrapper.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
@@ -36,7 +34,11 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEFixedPoint.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 
diff --git a/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.cpp b/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.cpp
index 8c11574..abaaf12 100644
--- a/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.cpp
+++ b/src/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.cpp
@@ -23,8 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -32,9 +30,13 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/Traits.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEElementwiseOperationKernel.cpp b/src/core/NEON/kernels/NEElementwiseOperationKernel.cpp
index f862d04..efe6161 100644
--- a/src/core/NEON/kernels/NEElementwiseOperationKernel.cpp
+++ b/src/core/NEON/kernels/NEElementwiseOperationKernel.cpp
@@ -23,12 +23,14 @@
  */
 #include "arm_compute/core/NEON/kernels/NEElementwiseOperationKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <map>
diff --git a/src/core/NEON/kernels/NEElementwiseUnaryKernel.cpp b/src/core/NEON/kernels/NEElementwiseUnaryKernel.cpp
index 40430bd..8e4b7ed 100644
--- a/src/core/NEON/kernels/NEElementwiseUnaryKernel.cpp
+++ b/src/core/NEON/kernels/NEElementwiseUnaryKernel.cpp
@@ -23,12 +23,14 @@
  */
 #include "arm_compute/core/NEON/kernels/NEElementwiseUnaryKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/ToolchainSupport.h"
 
 namespace arm_compute
diff --git a/src/core/NEON/kernels/NEErodeKernel.cpp b/src/core/NEON/kernels/NEErodeKernel.cpp
index 4b93c3b..31b0f48 100644
--- a/src/core/NEON/kernels/NEErodeKernel.cpp
+++ b/src/core/NEON/kernels/NEErodeKernel.cpp
@@ -28,6 +28,8 @@
 #include "arm_compute/core/NEON/INEKernel.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEFFTDigitReverseKernel.cpp b/src/core/NEON/kernels/NEFFTDigitReverseKernel.cpp
index d5b20d2..d8036f2 100644
--- a/src/core/NEON/kernels/NEFFTDigitReverseKernel.cpp
+++ b/src/core/NEON/kernels/NEFFTDigitReverseKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -28,6 +28,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <set>
 
diff --git a/src/core/NEON/kernels/NEFFTRadixStageKernel.cpp b/src/core/NEON/kernels/NEFFTRadixStageKernel.cpp
index de8ba3f..1b0af48 100644
--- a/src/core/NEON/kernels/NEFFTRadixStageKernel.cpp
+++ b/src/core/NEON/kernels/NEFFTRadixStageKernel.cpp
@@ -28,15 +28,16 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/NEON/wrapper/traits.h"
+#include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cmath>
 #include <complex>
 #include <map>
 
-#include "src/core/NEON/wrapper/traits.h"
-#include "src/core/NEON/wrapper/wrapper.h"
-
 namespace arm_compute
 {
 namespace
diff --git a/src/core/NEON/kernels/NEFFTScaleKernel.cpp b/src/core/NEON/kernels/NEFFTScaleKernel.cpp
index d99ff95..0cb8b84 100644
--- a/src/core/NEON/kernels/NEFFTScaleKernel.cpp
+++ b/src/core/NEON/kernels/NEFFTScaleKernel.cpp
@@ -29,6 +29,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEFastCornersKernel.cpp b/src/core/NEON/kernels/NEFastCornersKernel.cpp
index 7b1d81e..99312f5 100644
--- a/src/core/NEON/kernels/NEFastCornersKernel.cpp
+++ b/src/core/NEON/kernels/NEFastCornersKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -27,6 +27,8 @@
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NEFillArrayKernel.cpp b/src/core/NEON/kernels/NEFillArrayKernel.cpp
index 6b22dad..93798db 100644
--- a/src/core/NEON/kernels/NEFillArrayKernel.cpp
+++ b/src/core/NEON/kernels/NEFillArrayKernel.cpp
@@ -28,6 +28,7 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute;
 
diff --git a/src/core/NEON/kernels/NEFillBorderKernel.cpp b/src/core/NEON/kernels/NEFillBorderKernel.cpp
index dbaec83..c1dd5cf 100644
--- a/src/core/NEON/kernels/NEFillBorderKernel.cpp
+++ b/src/core/NEON/kernels/NEFillBorderKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <cstdint>
diff --git a/src/core/NEON/kernels/NEFlattenLayerKernel.cpp b/src/core/NEON/kernels/NEFlattenLayerKernel.cpp
index 35ebc5b..e6b34b6 100644
--- a/src/core/NEON/kernels/NEFlattenLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEFlattenLayerKernel.cpp
@@ -23,13 +23,15 @@
  */
 #include "arm_compute/core/NEON/kernels/NEFlattenLayerKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 
diff --git a/src/core/NEON/kernels/NEFloorKernel.cpp b/src/core/NEON/kernels/NEFloorKernel.cpp
index 301dc7a..48f964c 100644
--- a/src/core/NEON/kernels/NEFloorKernel.cpp
+++ b/src/core/NEON/kernels/NEFloorKernel.cpp
@@ -23,12 +23,14 @@
  */
 #include "arm_compute/core/NEON/kernels/NEFloorKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Coordinates.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/NEON/INEKernel.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "src/core/NEON/kernels/floor/impl/list.h"
 #include "src/core/common/Registrars.h"
diff --git a/src/core/NEON/kernels/NEFuseBatchNormalizationKernel.cpp b/src/core/NEON/kernels/NEFuseBatchNormalizationKernel.cpp
index 00d251f..e353df1 100644
--- a/src/core/NEON/kernels/NEFuseBatchNormalizationKernel.cpp
+++ b/src/core/NEON/kernels/NEFuseBatchNormalizationKernel.cpp
@@ -23,14 +23,16 @@
  */
 #include "arm_compute/core/NEON/kernels/NEFuseBatchNormalizationKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <map>
 
diff --git a/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.cpp b/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.cpp
index 8b4ad0d..2997c1d 100644
--- a/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.cpp
@@ -31,6 +31,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEGEMMLowpMatrixMultiplyKernel.cpp b/src/core/NEON/kernels/NEGEMMLowpMatrixMultiplyKernel.cpp
index f3ba290..acc5190 100644
--- a/src/core/NEON/kernels/NEGEMMLowpMatrixMultiplyKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMLowpMatrixMultiplyKernel.cpp
@@ -31,6 +31,10 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+
 #include <arm_neon.h>
 
 using namespace arm_compute;
@@ -1052,5 +1056,3 @@
     }
 }
 } // namespace arm_compute
-
-
diff --git a/src/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.cpp b/src/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.cpp
index 4ac33d1..1c76926 100644
--- a/src/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -32,6 +31,9 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEGEMMLowpOffsetContributionOutputStageKernel.cpp b/src/core/NEON/kernels/NEGEMMLowpOffsetContributionOutputStageKernel.cpp
index 8d0d7c2..6a7d225 100644
--- a/src/core/NEON/kernels/NEGEMMLowpOffsetContributionOutputStageKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMLowpOffsetContributionOutputStageKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGEMMLowpOffsetContributionOutputStageKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -32,8 +31,11 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ScaleKernel.cpp b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ScaleKernel.cpp
index 023b798..659c410 100644
--- a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ScaleKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ScaleKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ScaleKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -32,7 +31,10 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/AccessWindowStatic.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp
index 68f16c5..afa8cec 100644
--- a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -33,7 +32,10 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
 #include "src/core/NEON/NESymm.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp
index 2ef32c4..83416e0 100644
--- a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -33,7 +32,10 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
 #include "src/core/NEON/NEAsymm.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp
index 8fc33dc..1e8aa0c 100644
--- a/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -33,7 +32,10 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
 #include "src/core/NEON/NEAsymm.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEGEMMLowpReductionKernel.cpp b/src/core/NEON/kernels/NEGEMMLowpReductionKernel.cpp
index 1494cd4..566872f 100644
--- a/src/core/NEON/kernels/NEGEMMLowpReductionKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMLowpReductionKernel.cpp
@@ -23,11 +23,13 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGEMMLowpReductionKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/KernelDescriptors.h"
 #include "arm_compute/core/TensorInfo.h"
+#include "src/core/AccessWindowStatic.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEGEMMMatrixAdditionKernel.cpp b/src/core/NEON/kernels/NEGEMMMatrixAdditionKernel.cpp
index bd93146..9aee26c 100644
--- a/src/core/NEON/kernels/NEGEMMMatrixAdditionKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMMatrixAdditionKernel.cpp
@@ -23,12 +23,14 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEFixedPoint.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEGEMMMatrixMultiplyKernel.cpp b/src/core/NEON/kernels/NEGEMMMatrixMultiplyKernel.cpp
index 6f74e3f..a923689 100644
--- a/src/core/NEON/kernels/NEGEMMMatrixMultiplyKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMMatrixMultiplyKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGEMMMatrixMultiplyKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -32,9 +31,13 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/helpers/float_ops.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEFixedPoint.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/utils/helpers/float_ops.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp b/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp
index a8adc45..b9b4fe9 100644
--- a/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp
@@ -23,12 +23,14 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGEMMTranspose1xWKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/NEON/INEKernel.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEGatherKernel.cpp b/src/core/NEON/kernels/NEGatherKernel.cpp
index 906e8a0..193fe98 100644
--- a/src/core/NEON/kernels/NEGatherKernel.cpp
+++ b/src/core/NEON/kernels/NEGatherKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGatherKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Coordinates.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
@@ -32,6 +31,9 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEGaussian3x3Kernel.cpp b/src/core/NEON/kernels/NEGaussian3x3Kernel.cpp
index 18dd80e..5ff5db7 100644
--- a/src/core/NEON/kernels/NEGaussian3x3Kernel.cpp
+++ b/src/core/NEON/kernels/NEGaussian3x3Kernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -28,6 +28,8 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/NEON/INEKernel.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEGaussian5x5Kernel.cpp b/src/core/NEON/kernels/NEGaussian5x5Kernel.cpp
index 99b5d4b..5bb3e76 100644
--- a/src/core/NEON/kernels/NEGaussian5x5Kernel.cpp
+++ b/src/core/NEON/kernels/NEGaussian5x5Kernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEGaussianPyramidKernel.cpp b/src/core/NEON/kernels/NEGaussianPyramidKernel.cpp
index 83d2877..62cf414 100644
--- a/src/core/NEON/kernels/NEGaussianPyramidKernel.cpp
+++ b/src/core/NEON/kernels/NEGaussianPyramidKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -32,6 +32,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEGenerateProposalsLayerKernel.cpp b/src/core/NEON/kernels/NEGenerateProposalsLayerKernel.cpp
index c3b1059..483f204 100644
--- a/src/core/NEON/kernels/NEGenerateProposalsLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEGenerateProposalsLayerKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,12 +23,14 @@
  */
 #include "arm_compute/core/NEON/kernels/NEGenerateProposalsLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEHOGDescriptorKernel.cpp b/src/core/NEON/kernels/NEHOGDescriptorKernel.cpp
index 84bb59e..00f4087 100644
--- a/src/core/NEON/kernels/NEHOGDescriptorKernel.cpp
+++ b/src/core/NEON/kernels/NEHOGDescriptorKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -28,6 +28,8 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NEHOGDetectorKernel.cpp b/src/core/NEON/kernels/NEHOGDetectorKernel.cpp
index eb0d450..d5dfa41 100644
--- a/src/core/NEON/kernels/NEHOGDetectorKernel.cpp
+++ b/src/core/NEON/kernels/NEHOGDetectorKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -28,6 +28,8 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEHarrisCornersKernel.cpp b/src/core/NEON/kernels/NEHarrisCornersKernel.cpp
index 340c694..be68b9c 100644
--- a/src/core/NEON/kernels/NEHarrisCornersKernel.cpp
+++ b/src/core/NEON/kernels/NEHarrisCornersKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -31,6 +31,8 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NEHeightConcatenateLayerKernel.cpp b/src/core/NEON/kernels/NEHeightConcatenateLayerKernel.cpp
index fc7b819..a507125 100644
--- a/src/core/NEON/kernels/NEHeightConcatenateLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEHeightConcatenateLayerKernel.cpp
@@ -33,6 +33,8 @@
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <cstdint>
 
diff --git a/src/core/NEON/kernels/NEHistogramKernel.cpp b/src/core/NEON/kernels/NEHistogramKernel.cpp
index 0f8397f..12d1bb8 100644
--- a/src/core/NEON/kernels/NEHistogramKernel.cpp
+++ b/src/core/NEON/kernels/NEHistogramKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NEIm2ColKernel.cpp b/src/core/NEON/kernels/NEIm2ColKernel.cpp
index 6eae054..915ea75 100644
--- a/src/core/NEON/kernels/NEIm2ColKernel.cpp
+++ b/src/core/NEON/kernels/NEIm2ColKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEIm2ColKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -31,6 +30,9 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 
diff --git a/src/core/NEON/kernels/NEInstanceNormalizationLayerKernel.cpp b/src/core/NEON/kernels/NEInstanceNormalizationLayerKernel.cpp
index 78acbc3..7aa23de 100644
--- a/src/core/NEON/kernels/NEInstanceNormalizationLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEInstanceNormalizationLayerKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEInstanceNormalizationLayerKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -32,8 +31,11 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEIntegralImageKernel.cpp b/src/core/NEON/kernels/NEIntegralImageKernel.cpp
index 58ee3b4..5fc6ca6 100644
--- a/src/core/NEON/kernels/NEIntegralImageKernel.cpp
+++ b/src/core/NEON/kernels/NEIntegralImageKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -28,6 +28,8 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEL2NormalizeLayerKernel.cpp b/src/core/NEON/kernels/NEL2NormalizeLayerKernel.cpp
index d99def5..a216981 100644
--- a/src/core/NEON/kernels/NEL2NormalizeLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEL2NormalizeLayerKernel.cpp
@@ -31,6 +31,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/NEMath.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "src/core/NEON/wrapper/wrapper.h"
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NELKTrackerKernel.cpp b/src/core/NEON/kernels/NELKTrackerKernel.cpp
index 533c241..6567a8d 100644
--- a/src/core/NEON/kernels/NELKTrackerKernel.cpp
+++ b/src/core/NEON/kernels/NELKTrackerKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NELKTrackerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Coordinates.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
@@ -31,6 +30,9 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cmath>
diff --git a/src/core/NEON/kernels/NELocallyConnectedMatrixMultiplyKernel.cpp b/src/core/NEON/kernels/NELocallyConnectedMatrixMultiplyKernel.cpp
index 9eafe18..b8e6a6d 100644
--- a/src/core/NEON/kernels/NELocallyConnectedMatrixMultiplyKernel.cpp
+++ b/src/core/NEON/kernels/NELocallyConnectedMatrixMultiplyKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NELocallyConnectedMatrixMultiplyKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
@@ -33,7 +32,10 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEFixedPoint.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NEMagnitudePhaseKernel.cpp b/src/core/NEON/kernels/NEMagnitudePhaseKernel.cpp
index a0c1dbc..8d82e1a 100644
--- a/src/core/NEON/kernels/NEMagnitudePhaseKernel.cpp
+++ b/src/core/NEON/kernels/NEMagnitudePhaseKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -28,6 +28,8 @@
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstdint>
diff --git a/src/core/NEON/kernels/NEMaxUnpoolingLayerKernel.cpp b/src/core/NEON/kernels/NEMaxUnpoolingLayerKernel.cpp
index 821bf53..87caf00 100644
--- a/src/core/NEON/kernels/NEMaxUnpoolingLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEMaxUnpoolingLayerKernel.cpp
@@ -23,11 +23,13 @@
  */
 #include "arm_compute/core/NEON/kernels/NEMaxUnpoolingLayerKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "support/ToolchainSupport.h"
 
diff --git a/src/core/NEON/kernels/NEMeanStdDevKernel.cpp b/src/core/NEON/kernels/NEMeanStdDevKernel.cpp
index 914a21c..c4e036a 100644
--- a/src/core/NEON/kernels/NEMeanStdDevKernel.cpp
+++ b/src/core/NEON/kernels/NEMeanStdDevKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,6 +29,8 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cmath>
diff --git a/src/core/NEON/kernels/NEMeanStdDevNormalizationKernel.cpp b/src/core/NEON/kernels/NEMeanStdDevNormalizationKernel.cpp
index bcce843..8ee9ff6 100644
--- a/src/core/NEON/kernels/NEMeanStdDevNormalizationKernel.cpp
+++ b/src/core/NEON/kernels/NEMeanStdDevNormalizationKernel.cpp
@@ -23,14 +23,16 @@
  */
 #include "arm_compute/core/NEON/kernels/NEMeanStdDevNormalizationKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEMedian3x3Kernel.cpp b/src/core/NEON/kernels/NEMedian3x3Kernel.cpp
index 72225a4..86fcc30 100644
--- a/src/core/NEON/kernels/NEMedian3x3Kernel.cpp
+++ b/src/core/NEON/kernels/NEMedian3x3Kernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,6 +29,8 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/NEON/INEKernel.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <utility>
diff --git a/src/core/NEON/kernels/NEMemsetKernel.cpp b/src/core/NEON/kernels/NEMemsetKernel.cpp
index 3870fa5..fd427cc 100644
--- a/src/core/NEON/kernels/NEMemsetKernel.cpp
+++ b/src/core/NEON/kernels/NEMemsetKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,12 +23,14 @@
  */
 #include "arm_compute/core/NEON/kernels/NEMemsetKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEMinMaxLayerKernel.cpp b/src/core/NEON/kernels/NEMinMaxLayerKernel.cpp
index b1c2b1c..f675c39 100644
--- a/src/core/NEON/kernels/NEMinMaxLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEMinMaxLayerKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -33,6 +33,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NEMinMaxLocationKernel.cpp b/src/core/NEON/kernels/NEMinMaxLocationKernel.cpp
index e956f9a..e1691dc 100644
--- a/src/core/NEON/kernels/NEMinMaxLocationKernel.cpp
+++ b/src/core/NEON/kernels/NEMinMaxLocationKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -33,6 +33,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/Utility.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NENonLinearFilterKernel.cpp b/src/core/NEON/kernels/NENonLinearFilterKernel.cpp
index f20e869..31919ea 100644
--- a/src/core/NEON/kernels/NENonLinearFilterKernel.cpp
+++ b/src/core/NEON/kernels/NENonLinearFilterKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,6 +29,8 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <algorithm>
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NENonMaximaSuppression3x3Kernel.cpp b/src/core/NEON/kernels/NENonMaximaSuppression3x3Kernel.cpp
index 3e4c6e2..9566ced 100644
--- a/src/core/NEON/kernels/NENonMaximaSuppression3x3Kernel.cpp
+++ b/src/core/NEON/kernels/NENonMaximaSuppression3x3Kernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NENormalizationLayerKernel.cpp b/src/core/NEON/kernels/NENormalizationLayerKernel.cpp
index 7b88826..1b72a3e 100644
--- a/src/core/NEON/kernels/NENormalizationLayerKernel.cpp
+++ b/src/core/NEON/kernels/NENormalizationLayerKernel.cpp
@@ -23,17 +23,20 @@
  */
 #include "arm_compute/core/NEON/kernels/NENormalizationLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/NormalizationHelpers.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEPadLayerKernel.cpp b/src/core/NEON/kernels/NEPadLayerKernel.cpp
index 1b52117..ca9c541 100644
--- a/src/core/NEON/kernels/NEPadLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEPadLayerKernel.cpp
@@ -31,6 +31,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEPermuteKernel.cpp b/src/core/NEON/kernels/NEPermuteKernel.cpp
index 3f447f9..eab11eb 100644
--- a/src/core/NEON/kernels/NEPermuteKernel.cpp
+++ b/src/core/NEON/kernels/NEPermuteKernel.cpp
@@ -30,10 +30,12 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace
 {
-#include "arm_compute/core/NEON/kernels/convolution/common/shims.hpp"
+#include "src/core/NEON/kernels/convolution/common/shims.hpp"
 } // namespace
 
 namespace arm_compute
diff --git a/src/core/NEON/kernels/NEPixelWiseMultiplicationKernel.cpp b/src/core/NEON/kernels/NEPixelWiseMultiplicationKernel.cpp
index c5320b9..0847cb1 100644
--- a/src/core/NEON/kernels/NEPixelWiseMultiplicationKernel.cpp
+++ b/src/core/NEON/kernels/NEPixelWiseMultiplicationKernel.cpp
@@ -23,11 +23,13 @@
  */
 #include "arm_compute/core/NEON/kernels/NEPixelWiseMultiplicationKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/TensorInfo.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NESymm.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEPoolingLayerKernel.cpp b/src/core/NEON/kernels/NEPoolingLayerKernel.cpp
index 397eae9..f9636dc 100644
--- a/src/core/NEON/kernels/NEPoolingLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEPoolingLayerKernel.cpp
@@ -23,8 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEPoolingLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -33,9 +31,13 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/NEMath.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/ToolchainSupport.h"
 
 #include "src/core/NEON/wrapper/wrapper.h"
diff --git a/src/core/NEON/kernels/NEPriorBoxLayerKernel.cpp b/src/core/NEON/kernels/NEPriorBoxLayerKernel.cpp
index 808b68a..06a1f14 100644
--- a/src/core/NEON/kernels/NEPriorBoxLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEPriorBoxLayerKernel.cpp
@@ -27,6 +27,8 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEQLSTMLayerNormalizationKernel.cpp b/src/core/NEON/kernels/NEQLSTMLayerNormalizationKernel.cpp
index 6a038f8..55585b4 100644
--- a/src/core/NEON/kernels/NEQLSTMLayerNormalizationKernel.cpp
+++ b/src/core/NEON/kernels/NEQLSTMLayerNormalizationKernel.cpp
@@ -23,16 +23,18 @@
  */
 #include "arm_compute/core/NEON/kernels/NEQLSTMLayerNormalizationKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/NESymm.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "src/core/NEON/kernels/detail/NEActivationFunctionDetail.h"
 
diff --git a/src/core/NEON/kernels/NEQuantizationLayerKernel.cpp b/src/core/NEON/kernels/NEQuantizationLayerKernel.cpp
index 6d5202d..990e4b6 100644
--- a/src/core/NEON/kernels/NEQuantizationLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEQuantizationLayerKernel.cpp
@@ -31,8 +31,10 @@
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
-#include "arm_compute/core/CPP/Validate.h"
+#include "src/core/CPP/Validate.h"
 
 #include <arm_neon.h>
 #include <map>
diff --git a/src/core/NEON/kernels/NEROIAlignLayerKernel.cpp b/src/core/NEON/kernels/NEROIAlignLayerKernel.cpp
index 955cdc2..79f7888 100644
--- a/src/core/NEON/kernels/NEROIAlignLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEROIAlignLayerKernel.cpp
@@ -23,14 +23,16 @@
  */
 #include "arm_compute/core/NEON/kernels/NEROIAlignLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/misc/Utility.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEROIPoolingLayerKernel.cpp b/src/core/NEON/kernels/NEROIPoolingLayerKernel.cpp
index 6a960c7..a3171d9 100644
--- a/src/core/NEON/kernels/NEROIPoolingLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEROIPoolingLayerKernel.cpp
@@ -23,10 +23,12 @@
  */
 #include "arm_compute/core/NEON/kernels/NEROIPoolingLayerKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/ToolchainSupport.h"
 
 #include <cfloat>
diff --git a/src/core/NEON/kernels/NERangeKernel.cpp b/src/core/NEON/kernels/NERangeKernel.cpp
index 7d8fbb1..3466794 100644
--- a/src/core/NEON/kernels/NERangeKernel.cpp
+++ b/src/core/NEON/kernels/NERangeKernel.cpp
@@ -31,6 +31,8 @@
 #include "arm_compute/core/Validate.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "arm_compute/core/Utils.h"
 
diff --git a/src/core/NEON/kernels/NEReductionOperationKernel.cpp b/src/core/NEON/kernels/NEReductionOperationKernel.cpp
index 9af7f2a..716b092 100644
--- a/src/core/NEON/kernels/NEReductionOperationKernel.cpp
+++ b/src/core/NEON/kernels/NEReductionOperationKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEReductionOperationKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Coordinates.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
@@ -32,9 +31,12 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/utils/misc/SaturateCast.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEMath.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/SaturateCast.h"
 
 #include "src/core/NEON/wrapper/wrapper.h"
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NERemapKernel.cpp b/src/core/NEON/kernels/NERemapKernel.cpp
index 2881161..f698439 100644
--- a/src/core/NEON/kernels/NERemapKernel.cpp
+++ b/src/core/NEON/kernels/NERemapKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,13 +23,16 @@
  */
 #include "arm_compute/core/NEON/kernels/NERemapKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/ScaleHelpers.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
@@ -175,6 +178,8 @@
 
 void NERemapKernel::remap_bilinear(const Window &window)
 {
+    using namespace scale_helpers;
+
     // Don't increment in X and Y direction for the input tensor
     // A pointer to the start of this plane is needed as base for the precomputed offsets
     Window win_in(window);
diff --git a/src/core/NEON/kernels/NEReorgLayerKernel.cpp b/src/core/NEON/kernels/NEReorgLayerKernel.cpp
index 317bc25..1c48a5c 100644
--- a/src/core/NEON/kernels/NEReorgLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEReorgLayerKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <cstddef>
 #include <cstdint>
diff --git a/src/core/NEON/kernels/NEReshapeLayerKernel.cpp b/src/core/NEON/kernels/NEReshapeLayerKernel.cpp
index 23b349b..7946812 100644
--- a/src/core/NEON/kernels/NEReshapeLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEReshapeLayerKernel.cpp
@@ -23,8 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEReshapeLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
@@ -33,6 +31,10 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <cstdint>
 
diff --git a/src/core/NEON/kernels/NEReverseKernel.cpp b/src/core/NEON/kernels/NEReverseKernel.cpp
index 0c44a7e..2c081cb 100644
--- a/src/core/NEON/kernels/NEReverseKernel.cpp
+++ b/src/core/NEON/kernels/NEReverseKernel.cpp
@@ -27,6 +27,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEScaleKernel.cpp b/src/core/NEON/kernels/NEScaleKernel.cpp
index 94f5a18..e07fcad 100644
--- a/src/core/NEON/kernels/NEScaleKernel.cpp
+++ b/src/core/NEON/kernels/NEScaleKernel.cpp
@@ -23,15 +23,17 @@
  */
 #include "arm_compute/core/NEON/kernels/NEScaleKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/misc/Rounding.h"
 #include "arm_compute/core/utils/misc/Utility.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/ScaleHelpers.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "src/core/utils/ScaleUtils.h"
+#include "support/Rounding.h"
 
 #include <arm_neon.h>
 #include <map>
@@ -336,6 +338,8 @@
 
 void NEScaleKernel::scale_area_nchw_u8(const Window &window)
 {
+    using namespace scale_helpers;
+
     ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(_input, 1, DataType::U8);
 
     // Don't increment in width/height/channels for the input tensor
diff --git a/src/core/NEON/kernels/NEScharr3x3Kernel.cpp b/src/core/NEON/kernels/NEScharr3x3Kernel.cpp
index dcc9362..eb1dc65 100644
--- a/src/core/NEON/kernels/NEScharr3x3Kernel.cpp
+++ b/src/core/NEON/kernels/NEScharr3x3Kernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,6 +29,8 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstdint>
diff --git a/src/core/NEON/kernels/NESelectKernel.cpp b/src/core/NEON/kernels/NESelectKernel.cpp
index 286b8a6..2f36db2 100644
--- a/src/core/NEON/kernels/NESelectKernel.cpp
+++ b/src/core/NEON/kernels/NESelectKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NESelectKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
@@ -31,7 +30,10 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "utils/TypePrinter.h"
 
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NESobel3x3Kernel.cpp b/src/core/NEON/kernels/NESobel3x3Kernel.cpp
index eb9d3c3..1c7089b 100644
--- a/src/core/NEON/kernels/NESobel3x3Kernel.cpp
+++ b/src/core/NEON/kernels/NESobel3x3Kernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,6 +29,8 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstdint>
diff --git a/src/core/NEON/kernels/NESobel5x5Kernel.cpp b/src/core/NEON/kernels/NESobel5x5Kernel.cpp
index fc8ccc8..2421ea7 100644
--- a/src/core/NEON/kernels/NESobel5x5Kernel.cpp
+++ b/src/core/NEON/kernels/NESobel5x5Kernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/NESobel7x7Kernel.cpp b/src/core/NEON/kernels/NESobel7x7Kernel.cpp
index 95ab12b..779d67a 100644
--- a/src/core/NEON/kernels/NESobel7x7Kernel.cpp
+++ b/src/core/NEON/kernels/NESobel7x7Kernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 #include <cstdint>
diff --git a/src/core/NEON/kernels/NESoftmaxLayerKernel.cpp b/src/core/NEON/kernels/NESoftmaxLayerKernel.cpp
index e71818f..13f0a54 100644
--- a/src/core/NEON/kernels/NESoftmaxLayerKernel.cpp
+++ b/src/core/NEON/kernels/NESoftmaxLayerKernel.cpp
@@ -23,8 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NESoftmaxLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -32,10 +30,14 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/misc/SaturateCast.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "support/SaturateCast.h"
 
 #include <algorithm>
 #include <arm_neon.h>
diff --git a/src/core/NEON/kernels/NESpaceToBatchLayerKernel.cpp b/src/core/NEON/kernels/NESpaceToBatchLayerKernel.cpp
index ccad92a..3293466 100644
--- a/src/core/NEON/kernels/NESpaceToBatchLayerKernel.cpp
+++ b/src/core/NEON/kernels/NESpaceToBatchLayerKernel.cpp
@@ -29,6 +29,9 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+
 #include <arm_neon.h>
 #include <cstdint>
 
diff --git a/src/core/NEON/kernels/NESpaceToDepthLayerKernel.cpp b/src/core/NEON/kernels/NESpaceToDepthLayerKernel.cpp
index 2667611..7c9cc49 100644
--- a/src/core/NEON/kernels/NESpaceToDepthLayerKernel.cpp
+++ b/src/core/NEON/kernels/NESpaceToDepthLayerKernel.cpp
@@ -29,6 +29,9 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+
 #include <arm_neon.h>
 #include <cstdint>
 
diff --git a/src/core/NEON/kernels/NEStackLayerKernel.cpp b/src/core/NEON/kernels/NEStackLayerKernel.cpp
index 1d44be6..ad7f1b1 100644
--- a/src/core/NEON/kernels/NEStackLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEStackLayerKernel.cpp
@@ -33,6 +33,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 using namespace arm_compute;
 using namespace arm_compute::misc::shape_calculator;
diff --git a/src/core/NEON/kernels/NEStridedSliceKernel.cpp b/src/core/NEON/kernels/NEStridedSliceKernel.cpp
index 243a60f..13b2cb5 100644
--- a/src/core/NEON/kernels/NEStridedSliceKernel.cpp
+++ b/src/core/NEON/kernels/NEStridedSliceKernel.cpp
@@ -23,16 +23,17 @@
  */
 #include "arm_compute/core/NEON/kernels/NEStridedSliceKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
-#include "arm_compute/core/Window.h"
-
 #include "arm_compute/core/Types.h"
-#include "arm_compute/core/utils/helpers/bit_ops.h"
+#include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/helpers/tensor_transform.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/utils/helpers/bit_ops.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEThresholdKernel.cpp b/src/core/NEON/kernels/NEThresholdKernel.cpp
index 9e8ec5c..aad440b 100644
--- a/src/core/NEON/kernels/NEThresholdKernel.cpp
+++ b/src/core/NEON/kernels/NEThresholdKernel.cpp
@@ -27,6 +27,8 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "src/core/NEON/wrapper/wrapper.h"
 
diff --git a/src/core/NEON/kernels/NETileKernel.cpp b/src/core/NEON/kernels/NETileKernel.cpp
index cc7655a..99651c8 100644
--- a/src/core/NEON/kernels/NETileKernel.cpp
+++ b/src/core/NEON/kernels/NETileKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NETransposeKernel.cpp b/src/core/NEON/kernels/NETransposeKernel.cpp
index 7118e45..6037810 100644
--- a/src/core/NEON/kernels/NETransposeKernel.cpp
+++ b/src/core/NEON/kernels/NETransposeKernel.cpp
@@ -23,14 +23,16 @@
  */
 #include "arm_compute/core/NEON/kernels/NETransposeKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/AccessWindowTranspose.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/AccessWindowTranspose.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEUpsampleLayerKernel.cpp b/src/core/NEON/kernels/NEUpsampleLayerKernel.cpp
index 69324c1..129c83c 100644
--- a/src/core/NEON/kernels/NEUpsampleLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEUpsampleLayerKernel.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEUpsampleLayerKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -31,7 +30,10 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/NEON/kernels/NEWarpKernel.cpp b/src/core/NEON/kernels/NEWarpKernel.cpp
index d8191dc..891304f 100644
--- a/src/core/NEON/kernels/NEWarpKernel.cpp
+++ b/src/core/NEON/kernels/NEWarpKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2018 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/core/NEON/kernels/NEWarpKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Coordinates.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
@@ -31,6 +30,10 @@
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/ScaleHelpers.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <cstddef>
 
@@ -184,7 +187,7 @@
                     *out.ptr() = nearest_interpolation(in.ptr(), x0, y0, stride);
                     break;
                 case InterpolationPolicy::BILINEAR:
-                    *out.ptr() = pixel_bilinear_c1(in.ptr(), stride, x0, y0);
+                    *out.ptr() = scale_helpers::pixel_bilinear_c1(in.ptr(), stride, x0, y0);
                     break;
                 default:
                     ARM_COMPUTE_ERROR("Interpolation not supported");
@@ -271,7 +274,7 @@
                     *out.ptr() = nearest_interpolation(in.ptr(), x0, y0, stride);
                     break;
                 case InterpolationPolicy::BILINEAR:
-                    *out.ptr() = pixel_bilinear_c1(in.ptr(), stride, x0, y0);
+                    *out.ptr() = scale_helpers::pixel_bilinear_c1(in.ptr(), stride, x0, y0);
                     break;
                 default:
                     ARM_COMPUTE_ERROR("Interpolation not supported");
@@ -386,7 +389,7 @@
                     *out.ptr() = nearest_interpolation(in.ptr(), x0, y0, stride);
                     break;
                 case InterpolationPolicy::BILINEAR:
-                    *out.ptr() = pixel_bilinear_c1(in.ptr(), stride, x0, y0);
+                    *out.ptr() = scale_helpers::pixel_bilinear_c1(in.ptr(), stride, x0, y0);
                     break;
                 default:
                     ARM_COMPUTE_ERROR("Interpolation not supported");
@@ -519,7 +522,7 @@
                     *out.ptr() = nearest_interpolation(in.ptr(), xn, yn, stride);
                     break;
                 case InterpolationPolicy::BILINEAR:
-                    *out.ptr() = pixel_bilinear_c1(in.ptr(), stride, xn, yn);
+                    *out.ptr() = scale_helpers::pixel_bilinear_c1(in.ptr(), stride, xn, yn);
                     break;
                 default:
                     ARM_COMPUTE_ERROR("Interpolation not supported");
@@ -620,7 +623,7 @@
                     *out.ptr() = nearest_interpolation(in.ptr(), xn, yn, stride);
                     break;
                 case InterpolationPolicy::BILINEAR:
-                    *out.ptr() = pixel_bilinear_c1(in.ptr(), stride, xn, yn);
+                    *out.ptr() = scale_helpers::pixel_bilinear_c1(in.ptr(), stride, xn, yn);
                     break;
                 default:
                     ARM_COMPUTE_ERROR("Interpolation not supported");
@@ -752,7 +755,7 @@
                     *out.ptr() = nearest_interpolation(in.ptr(), xn, yn, stride);
                     break;
                 case InterpolationPolicy::BILINEAR:
-                    *out.ptr() = pixel_bilinear_c1(in.ptr(), stride, xn, yn);
+                    *out.ptr() = scale_helpers::pixel_bilinear_c1(in.ptr(), stride, xn, yn);
                     break;
                 default:
                     ARM_COMPUTE_ERROR("Interpolation not supported");
diff --git a/src/core/NEON/kernels/NEWeightsReshapeKernel.cpp b/src/core/NEON/kernels/NEWeightsReshapeKernel.cpp
index 6a74914..c7fa2d2 100644
--- a/src/core/NEON/kernels/NEWeightsReshapeKernel.cpp
+++ b/src/core/NEON/kernels/NEWeightsReshapeKernel.cpp
@@ -25,6 +25,8 @@
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/core/NEON/kernels/NEWidthConcatenateLayerKernel.cpp b/src/core/NEON/kernels/NEWidthConcatenateLayerKernel.cpp
index d12b10c..90afbd6 100644
--- a/src/core/NEON/kernels/NEWidthConcatenateLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEWidthConcatenateLayerKernel.cpp
@@ -33,6 +33,8 @@
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include <cstdint>
 
diff --git a/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp b/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp
index bfe97bf..211ebde 100644
--- a/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp
@@ -23,16 +23,18 @@
  */
 #include "src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h"
 
-#include "arm_compute/core/AccessWindowStatic.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/IAccessWindow.h"
 #include "arm_compute/core/ITensor.h"
-#include "arm_compute/core/NEON/kernels/convolution/common/utils.hpp"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/AccessWindowStatic.h"
+#include "src/core/NEON/kernels/convolution/common/utils.hpp"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 #include "support/MemorySupport.h"
 
 #include "src/core/NEON/kernels/convolution/winograd/winograd_layer.hpp"
diff --git a/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h b/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h
index 94df4f6..bf5d77f 100644
--- a/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h
+++ b/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_NEGEMMWINOGRADCONVOLUTIONLAYERKERNEL_H
 
 #include "arm_compute/core/NEON/INEKernel.h"
-#include "arm_compute/core/NEON/kernels/convolution/common/convolution.hpp"
-#include "arm_compute/core/NEON/kernels/convolution/common/tensor.hpp"
+#include "src/core/NEON/kernels/convolution/common/convolution.hpp"
+#include "src/core/NEON/kernels/convolution/common/tensor.hpp"
 
 #include "src/core/NEON/kernels/convolution/winograd/winograd_layer.hpp"
 
diff --git a/src/core/NEON/kernels/NEYOLOLayerKernel.cpp b/src/core/NEON/kernels/NEYOLOLayerKernel.cpp
index 591aa1e..48c0616 100644
--- a/src/core/NEON/kernels/NEYOLOLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEYOLOLayerKernel.cpp
@@ -23,16 +23,18 @@
  */
 #include "arm_compute/core/NEON/kernels/NEYOLOLayerKernel.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/NEMath.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 #include "src/core/NEON/kernels/detail/NEActivationFunctionDetail.h"
 
diff --git a/src/core/NEON/kernels/assembly/INEGEMMWrapperKernel.cpp b/src/core/NEON/kernels/assembly/INEGEMMWrapperKernel.cpp
index b071be3..760274d 100644
--- a/src/core/NEON/kernels/assembly/INEGEMMWrapperKernel.cpp
+++ b/src/core/NEON/kernels/assembly/INEGEMMWrapperKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -22,7 +22,7 @@
  * SOFTWARE.
  */
 
-#include "arm_compute/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h"
+#include "src/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h"
 
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
diff --git a/arm_compute/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h b/src/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h
similarity index 96%
rename from arm_compute/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h
rename to src/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h
index 74161e3..030f1aa 100644
--- a/arm_compute/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h
+++ b/src/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,8 +21,8 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_INEGEMMWRAPPERKERNEL_H
-#define ARM_COMPUTE_INEGEMMWRAPPERKERNEL_H
+#ifndef SRC_INEGEMMWRAPPERKERNEL_H
+#define SRC_INEGEMMWRAPPERKERNEL_H
 
 #include "arm_compute/core/NEON/INEKernel.h"
 
@@ -105,4 +105,4 @@
 
 } // namespace arm_compute
 
-#endif /* ARM_COMPUTE_INEGEMMRAPPERKERNEL_H */
+#endif /* SRC_INEGEMMRAPPERKERNEL_H */
diff --git a/arm_compute/core/NEON/kernels/assembly/NEDepthwiseConvolutionAssemblyKernelWrapper.h b/src/core/NEON/kernels/assembly/NEDepthwiseConvolutionAssemblyKernelWrapper.h
similarity index 90%
rename from arm_compute/core/NEON/kernels/assembly/NEDepthwiseConvolutionAssemblyKernelWrapper.h
rename to src/core/NEON/kernels/assembly/NEDepthwiseConvolutionAssemblyKernelWrapper.h
index 7c10f85..a2f7e3b 100644
--- a/arm_compute/core/NEON/kernels/assembly/NEDepthwiseConvolutionAssemblyKernelWrapper.h
+++ b/src/core/NEON/kernels/assembly/NEDepthwiseConvolutionAssemblyKernelWrapper.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,14 +21,14 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_ASSEMBLY_DEPTHWISE_CONVOLUTION_ASSEMBLY_WRAPPER_KERNEL_H
-#define ARM_COMPUTE_ASSEMBLY_DEPTHWISE_CONVOLUTION_ASSEMBLY_WRAPPER_KERNEL_H
+#ifndef SRC_ASSEMBLY_DEPTHWISE_CONVOLUTION_ASSEMBLY_WRAPPER_KERNEL_H
+#define SRC_ASSEMBLY_DEPTHWISE_CONVOLUTION_ASSEMBLY_WRAPPER_KERNEL_H
 
 #include "arm_compute/core/NEON/INEKernel.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 
-#include "arm_compute/core/NEON/kernels/convolution/depthwise/depthwise.hpp"
+#include "src/core/NEON/kernels/convolution/depthwise/depthwise.hpp"
 
 namespace arm_compute
 {
@@ -85,4 +85,4 @@
     depthwise::IDepthwiseConvolution *_kernel;
 };
 } // namespace arm_compute
-#endif /* ARM_COMPUTE_ASSEMBLY_DEPTHWISE_CONVOLUTION_ASSEMBLY_WRAPPER_KERNEL_H */
+#endif /* SRC_ASSEMBLY_DEPTHWISE_CONVOLUTION_ASSEMBLY_WRAPPER_KERNEL_H */
diff --git a/arm_compute/core/utils/misc/ICloneable.h b/src/core/NEON/kernels/assembly/arm_gemm_local.hpp
similarity index 67%
copy from arm_compute/core/utils/misc/ICloneable.h
copy to src/core/NEON/kernels/assembly/arm_gemm_local.hpp
index cbb0b3c..4715f25 100644
--- a/arm_compute/core/utils/misc/ICloneable.h
+++ b/src/core/NEON/kernels/assembly/arm_gemm_local.hpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,28 +21,14 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_MISC_ICLONEABLE_H
-#define ARM_COMPUTE_MISC_ICLONEABLE_H
+#pragma once
 
-#include <memory>
+/* This file is used to configure integration-specific aspects of arm_gemm into ACL */
 
-namespace arm_compute
+#include "arm_compute/core/CPP/CPPTypes.h"
+
+namespace arm_gemm
 {
-namespace misc
-{
-/** Clonable Interface */
-template <class T>
-class ICloneable
-{
-public:
-    /** Default virtual desctructor */
-    virtual ~ICloneable() = default;
-    /** Provide a clone of the current object of class T
-     *
-     * @return Clone object of class T
-     */
-    virtual std::unique_ptr<T> clone() const = 0;
-};
-} // namespace misc
+using CPUModel = arm_compute::CPUModel;
+using CPUInfo  = arm_compute::CPUInfo;
 } // namespace arm_compute
-#endif /* ARM_COMPUTE_MISC_ICLONEABLE_H */
diff --git a/arm_compute/core/NEON/kernels/convolution/common/activation.hpp b/src/core/NEON/kernels/convolution/common/activation.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/activation.hpp
rename to src/core/NEON/kernels/convolution/common/activation.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/common/alloc.hpp b/src/core/NEON/kernels/convolution/common/alloc.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/alloc.hpp
rename to src/core/NEON/kernels/convolution/common/alloc.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/common/arm.hpp b/src/core/NEON/kernels/convolution/common/arm.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/arm.hpp
rename to src/core/NEON/kernels/convolution/common/arm.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/common/convolution.hpp b/src/core/NEON/kernels/convolution/common/convolution.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/convolution.hpp
rename to src/core/NEON/kernels/convolution/common/convolution.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/common/padding.hpp b/src/core/NEON/kernels/convolution/common/padding.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/padding.hpp
rename to src/core/NEON/kernels/convolution/common/padding.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/common/perf.h b/src/core/NEON/kernels/convolution/common/perf.h
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/perf.h
rename to src/core/NEON/kernels/convolution/common/perf.h
diff --git a/arm_compute/core/NEON/kernels/convolution/common/qasymm8.hpp b/src/core/NEON/kernels/convolution/common/qasymm8.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/qasymm8.hpp
rename to src/core/NEON/kernels/convolution/common/qasymm8.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/common/qsymm8.hpp b/src/core/NEON/kernels/convolution/common/qsymm8.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/qsymm8.hpp
rename to src/core/NEON/kernels/convolution/common/qsymm8.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/common/shims.hpp b/src/core/NEON/kernels/convolution/common/shims.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/shims.hpp
rename to src/core/NEON/kernels/convolution/common/shims.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/common/tensor.hpp b/src/core/NEON/kernels/convolution/common/tensor.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/tensor.hpp
rename to src/core/NEON/kernels/convolution/common/tensor.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/common/tensor_utils.hpp b/src/core/NEON/kernels/convolution/common/tensor_utils.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/tensor_utils.hpp
rename to src/core/NEON/kernels/convolution/common/tensor_utils.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/common/utils.hpp b/src/core/NEON/kernels/convolution/common/utils.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/common/utils.hpp
rename to src/core/NEON/kernels/convolution/common/utils.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/depthwise/depthwise.hpp b/src/core/NEON/kernels/convolution/depthwise/depthwise.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/depthwise/depthwise.hpp
rename to src/core/NEON/kernels/convolution/depthwise/depthwise.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/depthwise/depthwise_dilated.hpp b/src/core/NEON/kernels/convolution/depthwise/depthwise_dilated.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/depthwise/depthwise_dilated.hpp
rename to src/core/NEON/kernels/convolution/depthwise/depthwise_dilated.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/depthwise/depthwise_quantized.hpp b/src/core/NEON/kernels/convolution/depthwise/depthwise_quantized.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/depthwise/depthwise_quantized.hpp
rename to src/core/NEON/kernels/convolution/depthwise/depthwise_quantized.hpp
diff --git a/arm_compute/core/NEON/kernels/convolution/depthwise/depthwise_quantized_dilated.hpp b/src/core/NEON/kernels/convolution/depthwise/depthwise_quantized_dilated.hpp
similarity index 100%
rename from arm_compute/core/NEON/kernels/convolution/depthwise/depthwise_quantized_dilated.hpp
rename to src/core/NEON/kernels/convolution/depthwise/depthwise_quantized_dilated.hpp
diff --git a/src/core/NEON/kernels/detail/NEDirectConvolutionDetail.h b/src/core/NEON/kernels/detail/NEDirectConvolutionDetail.h
index d7ee70a..59f5c6c 100644
--- a/src/core/NEON/kernels/detail/NEDirectConvolutionDetail.h
+++ b/src/core/NEON/kernels/detail/NEDirectConvolutionDetail.h
@@ -25,10 +25,10 @@
 #ifndef ARM_COMPUTE_NEDIRECTCONVOLUTIONDETAIL_H
 #define ARM_COMPUTE_NEDIRECTCONVOLUTIONDETAIL_H
 
-#include "arm_compute/core/AccessWindowStatic.h"
-#include "arm_compute/core/utils/misc/Requires.h"
+#include "src/core/AccessWindowStatic.h"
 #include "src/core/NEON/NEFixedPoint.h"
 #include "src/core/NEON/wrapper/wrapper.h"
+#include "support/Requires.h"
 
 #include <arm_neon.h>
 
diff --git a/src/core/TensorInfo.cpp b/src/core/TensorInfo.cpp
index c1a1c2e..414c128 100644
--- a/src/core/TensorInfo.cpp
+++ b/src/core/TensorInfo.cpp
@@ -28,6 +28,7 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/Utils.h"
 #include "support/MemorySupport.h"
 
 using namespace arm_compute;
diff --git a/src/core/helpers/AutoConfiguration.h b/src/core/helpers/AutoConfiguration.h
new file mode 100644
index 0000000..6880a6c
--- /dev/null
+++ b/src/core/helpers/AutoConfiguration.h
@@ -0,0 +1,176 @@
+/*
+* Copyright (c) 2020 Arm Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef SRC_CORE_HELPERS_AUTOCONFIGURATION_H
+#define SRC_CORE_HELPERS_AUTOCONFIGURATION_H
+
+#include "arm_compute/core/ITensorInfo.h"
+#include "arm_compute/core/Types.h"
+
+namespace arm_compute
+{
+/** Auto initialize the tensor info (shape, number of channels and data type) if the current assignment is empty.
+ *
+ * @param[in,out] info              Tensor info used to check and assign.
+ * @param[in]     shape             New shape.
+ * @param[in]     num_channels      New number of channels.
+ * @param[in]     data_type         New data type
+ * @param[in]     quantization_info (Optional) New quantization info
+ *
+ * @return True if the tensor info has been initialized
+ */
+inline bool auto_init_if_empty(ITensorInfo       &info,
+                               const TensorShape &shape,
+                               int num_channels, DataType data_type,
+                               QuantizationInfo quantization_info = QuantizationInfo())
+{
+    if(info.tensor_shape().total_size() == 0)
+    {
+        info.set_data_type(data_type);
+        info.set_num_channels(num_channels);
+        info.set_tensor_shape(shape);
+        info.set_quantization_info(quantization_info);
+        return true;
+    }
+
+    return false;
+}
+
+/** Auto initialize the tensor info using another tensor info.
+*
+* @param info_sink   Tensor info used to check and assign
+* @param info_source Tensor info used to assign
+*
+* @return True if the tensor info has been initialized
+*/
+inline bool auto_init_if_empty(ITensorInfo &info_sink, const ITensorInfo &info_source)
+{
+    if(info_sink.tensor_shape().total_size() == 0)
+    {
+        info_sink.set_data_type(info_source.data_type());
+        info_sink.set_num_channels(info_source.num_channels());
+        info_sink.set_tensor_shape(info_source.tensor_shape());
+        info_sink.set_quantization_info(info_source.quantization_info());
+        info_sink.set_data_layout(info_source.data_layout());
+        return true;
+    }
+
+    return false;
+}
+
+/** Set the shape to the specified value if the current assignment is empty.
+ *
+ * @param[in,out] info  Tensor info used to check and assign.
+ * @param[in]     shape New shape.
+ *
+ * @return True if the shape has been changed.
+ */
+inline bool set_shape_if_empty(ITensorInfo &info, const TensorShape &shape)
+{
+    if(info.tensor_shape().total_size() == 0)
+    {
+        info.set_tensor_shape(shape);
+        return true;
+    }
+
+    return false;
+}
+
+/** Set the format, data type and number of channels to the specified value if
+ * the current data type is unknown.
+ *
+ * @param[in,out] info   Tensor info used to check and assign.
+ * @param[in]     format New format.
+ *
+ * @return True if the format has been changed.
+ */
+inline bool set_format_if_unknown(ITensorInfo &info, Format format)
+{
+    if(info.data_type() == DataType::UNKNOWN)
+    {
+        info.set_format(format);
+        return true;
+    }
+
+    return false;
+}
+
+/** Set the data type and number of channels to the specified value if
+ * the current data type is unknown.
+ *
+ * @param[in,out] info      Tensor info used to check and assign.
+ * @param[in]     data_type New data type.
+ *
+ * @return True if the data type has been changed.
+ */
+inline bool set_data_type_if_unknown(ITensorInfo &info, DataType data_type)
+{
+    if(info.data_type() == DataType::UNKNOWN)
+    {
+        info.set_data_type(data_type);
+        return true;
+    }
+
+    return false;
+}
+
+/** Set the data layout to the specified value if
+ * the current data layout is unknown.
+ *
+ * @param[in,out] info        Tensor info used to check and assign.
+ * @param[in]     data_layout New data layout.
+ *
+ * @return True if the data type has been changed.
+ */
+inline bool set_data_layout_if_unknown(ITensorInfo &info, DataLayout data_layout)
+{
+    if(info.data_layout() == DataLayout::UNKNOWN)
+    {
+        info.set_data_layout(data_layout);
+        return true;
+    }
+
+    return false;
+}
+
+/** Set the quantization info to the specified value if
+ * the current quantization info is empty and the data type of asymmetric quantized type
+ *
+ * @param[in,out] info              Tensor info used to check and assign.
+ * @param[in]     quantization_info Quantization info
+ *
+ * @return True if the quantization info has been changed.
+ */
+inline bool set_quantization_info_if_empty(ITensorInfo &info, QuantizationInfo quantization_info)
+{
+    if(info.quantization_info().empty() && (is_data_type_quantized_asymmetric(info.data_type())))
+    {
+        info.set_quantization_info(quantization_info);
+        return true;
+    }
+
+    return false;
+}
+} // namespace arm_compute
+
+#endif /* SRC_CORE_HELPERS_AUTOCONFIGURATION_H */
diff --git a/src/core/helpers/NormalizationHelpers.h b/src/core/helpers/NormalizationHelpers.h
new file mode 100644
index 0000000..d94d5e3
--- /dev/null
+++ b/src/core/helpers/NormalizationHelpers.h
@@ -0,0 +1,47 @@
+/*
+* Copyright (c) 2020 Arm Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef SRC_CORE_HELPERS_NORMALIZATIONHELPERS_H
+#define SRC_CORE_HELPERS_NORMALIZATIONHELPERS_H
+
+#include "arm_compute/core/Helpers.h"
+#include "arm_compute/core/Types.h"
+
+namespace arm_compute
+{
+/** Calculate the normalization dimension index for a given normalization type
+ *
+ * @param[in] layout Data layout of the input and output tensor
+ * @param[in] info   Normalization info
+ *
+ * @return Normalization dimension index
+ */
+inline unsigned int get_normalization_dimension_index(DataLayout layout, const NormalizationLayerInfo &info)
+{
+    const unsigned int width_idx   = get_data_layout_dimension_index(layout, DataLayoutDimension::WIDTH);
+    const unsigned int channel_idx = get_data_layout_dimension_index(layout, DataLayoutDimension::CHANNEL);
+
+    return info.is_in_map() ? width_idx : channel_idx;
+}
+} // namespace arm_compute
+#endif /* SRC_CORE_HELPERS_NORMALIZATIONHELPERS_H */
diff --git a/src/core/helpers/ScaleHelpers.h b/src/core/helpers/ScaleHelpers.h
new file mode 100644
index 0000000..827bbef
--- /dev/null
+++ b/src/core/helpers/ScaleHelpers.h
@@ -0,0 +1,331 @@
+/*
+* Copyright (c) 2020 Arm Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef SRC_CORE_HELPERS_SCALEHELPERS_H
+#define SRC_CORE_HELPERS_SCALEHELPERS_H
+
+#include "arm_compute/core/Error.h"
+#include "arm_compute/core/QuantizationInfo.h"
+
+#include <algorithm>
+#include <cmath>
+#include <cstddef>
+#include <cstdint>
+
+namespace arm_compute
+{
+namespace scale_helpers
+{
+/** Computes bilinear interpolation using the pointer to the top-left pixel and the pixel's distance between
+ * the real coordinates and the smallest following integer coordinates. Input must be in single channel format.
+ *
+ * @param[in] pixel_ptr Pointer to the top-left pixel value of a single channel input.
+ * @param[in] stride    Stride to access the bottom-left and bottom-right pixel values
+ * @param[in] dx        Pixel's distance between the X real coordinate and the smallest X following integer
+ * @param[in] dy        Pixel's distance between the Y real coordinate and the smallest Y following integer
+ *
+ * @note dx and dy must be in the range [0, 1.0]
+ *
+ * @return The bilinear interpolated pixel value
+ */
+template <typename T>
+inline T delta_bilinear_c1(const T *pixel_ptr, size_t stride, float dx, float dy)
+{
+    ARM_COMPUTE_ERROR_ON(pixel_ptr == nullptr);
+
+    const float dx1 = 1.0f - dx;
+    const float dy1 = 1.0f - dy;
+
+    const T a00 = *pixel_ptr;
+    const T a01 = *(pixel_ptr + 1);
+    const T a10 = *(pixel_ptr + stride);
+    const T a11 = *(pixel_ptr + stride + 1);
+
+    const float w1 = dx1 * dy1;
+    const float w2 = dx * dy1;
+    const float w3 = dx1 * dy;
+    const float w4 = dx * dy;
+
+    return static_cast<T>(a00 * w1 + a01 * w2 + a10 * w3 + a11 * w4);
+}
+
+/** Computes bilinear interpolation for quantized input and output, using the pointer to the top-left pixel and the pixel's distance between
+ * the real coordinates and the smallest following integer coordinates. Input must be QASYMM8 and in single channel format.
+ *
+ * @param[in] pixel_ptr Pointer to the top-left pixel value of a single channel input.
+ * @param[in] stride    Stride to access the bottom-left and bottom-right pixel values
+ * @param[in] dx        Pixel's distance between the X real coordinate and the smallest X following integer
+ * @param[in] dy        Pixel's distance between the Y real coordinate and the smallest Y following integer
+ * @param[in] iq_info   Input QuantizationInfo
+ * @param[in] oq_info   Output QuantizationInfo
+ *
+ * @note dx and dy must be in the range [0, 1.0]
+ *
+ * @return The bilinear interpolated pixel value
+ */
+inline uint8_t delta_bilinear_c1_quantized(const uint8_t *pixel_ptr, size_t stride, float dx, float dy,
+                                           UniformQuantizationInfo iq_info, UniformQuantizationInfo oq_info)
+{
+    ARM_COMPUTE_ERROR_ON(pixel_ptr == nullptr);
+
+    const float dx1 = 1.0f - dx;
+    const float dy1 = 1.0f - dy;
+
+    const float a00 = dequantize_qasymm8(*pixel_ptr, iq_info);
+    const float a01 = dequantize_qasymm8(*(pixel_ptr + 1), iq_info);
+    const float a10 = dequantize_qasymm8(*(pixel_ptr + stride), iq_info);
+    const float a11 = dequantize_qasymm8(*(pixel_ptr + stride + 1), iq_info);
+
+    const float w1  = dx1 * dy1;
+    const float w2  = dx * dy1;
+    const float w3  = dx1 * dy;
+    const float w4  = dx * dy;
+    float       res = a00 * w1 + a01 * w2 + a10 * w3 + a11 * w4;
+    return static_cast<uint8_t>(quantize_qasymm8(res, oq_info));
+}
+
+/** Computes bilinear interpolation for quantized input and output, using the pointer to the top-left pixel and the pixel's distance between
+ * the real coordinates and the smallest following integer coordinates. Input must be QASYMM8_SIGNED and in single channel format.
+ *
+ * @param[in] pixel_ptr Pointer to the top-left pixel value of a single channel input.
+ * @param[in] stride    Stride to access the bottom-left and bottom-right pixel values
+ * @param[in] dx        Pixel's distance between the X real coordinate and the smallest X following integer
+ * @param[in] dy        Pixel's distance between the Y real coordinate and the smallest Y following integer
+ * @param[in] iq_info   Input QuantizationInfo
+ * @param[in] oq_info   Output QuantizationInfo
+ *
+ * @note dx and dy must be in the range [0, 1.0]
+ *
+ * @return The bilinear interpolated pixel value
+ */
+inline int8_t delta_bilinear_c1_quantized(const int8_t *pixel_ptr, size_t stride, float dx, float dy,
+                                          UniformQuantizationInfo iq_info, UniformQuantizationInfo oq_info)
+{
+    ARM_COMPUTE_ERROR_ON(pixel_ptr == nullptr);
+
+    const float dx1 = 1.0f - dx;
+    const float dy1 = 1.0f - dy;
+
+    const float a00 = dequantize_qasymm8_signed(*pixel_ptr, iq_info);
+    const float a01 = dequantize_qasymm8_signed(*(pixel_ptr + 1), iq_info);
+    const float a10 = dequantize_qasymm8_signed(*(pixel_ptr + stride), iq_info);
+    const float a11 = dequantize_qasymm8_signed(*(pixel_ptr + stride + 1), iq_info);
+
+    const float w1  = dx1 * dy1;
+    const float w2  = dx * dy1;
+    const float w3  = dx1 * dy;
+    const float w4  = dx * dy;
+    float       res = a00 * w1 + a01 * w2 + a10 * w3 + a11 * w4;
+    return static_cast<int8_t>(quantize_qasymm8_signed(res, oq_info));
+}
+
+/** Computes linear interpolation using the pointer to the top pixel and the pixel's distance between
+ * the real coordinates and the smallest following integer coordinates. Input must be in single channel format.
+ *
+ * @param[in] pixel_ptr Pointer to the top pixel value of a single channel input.
+ * @param[in] stride    Stride to access the bottom pixel value
+ * @param[in] dy        Pixel's distance between the Y real coordinate and the smallest Y following integer
+ *
+ * @note dy must be in the range [0, 1.0]
+ *
+ * @return The linear interpolated pixel value
+ */
+template <typename T>
+inline T delta_linear_c1_y(const T *pixel_ptr, size_t stride, float dy)
+{
+    ARM_COMPUTE_ERROR_ON(pixel_ptr == nullptr);
+
+    const float dy1 = 1.0f - dy;
+
+    const T a00 = *pixel_ptr;
+    const T a10 = *(pixel_ptr + stride);
+
+    const float w1 = dy1;
+    const float w3 = dy;
+
+    return static_cast<T>(a00 * w1 + a10 * w3);
+}
+
+/** Computes linear interpolation using the pointer to the left pixel and the pixel's distance between
+ * the real coordinates and the smallest following integer coordinates. Input must be in single channel format.
+ *
+ * @param[in] pixel_ptr Pointer to the left pixel value of a single channel input.
+ * @param[in] dx        Pixel's distance between the X real coordinate and the smallest X following integer
+ *
+ * @note dx must be in the range [0, 1.0]
+ *
+ * @return The linear interpolated pixel value
+ */
+template <typename T>
+inline T delta_linear_c1_x(const T *pixel_ptr, float dx)
+{
+    ARM_COMPUTE_ERROR_ON(pixel_ptr == nullptr);
+
+    const T a00 = *pixel_ptr;
+    const T a01 = *(pixel_ptr + 1);
+
+    const float dx1 = 1.0f - dx;
+
+    const float w1 = dx1;
+    const float w2 = dx;
+
+    return static_cast<T>(a00 * w1 + a01 * w2);
+}
+
+/** Return the pixel at (x,y) using bilinear interpolation.
+ *
+ * @warning Only works if the iterator was created with an IImage
+ *
+ * @param[in] first_pixel_ptr Pointer to the first pixel of a single channel input.
+ * @param[in] stride          Stride in bytes of the image;
+ * @param[in] x               X position of the wanted pixel
+ * @param[in] y               Y position of the wanted pixel
+ *
+ * @return The pixel at (x, y) using bilinear interpolation.
+ */
+template <typename T>
+inline T pixel_bilinear_c1(const T *first_pixel_ptr, size_t stride, float x, float y)
+{
+    ARM_COMPUTE_ERROR_ON(first_pixel_ptr == nullptr);
+
+    const int32_t xi = std::floor(x);
+    const int32_t yi = std::floor(y);
+
+    const float dx = x - xi;
+    const float dy = y - yi;
+
+    return delta_bilinear_c1(first_pixel_ptr + xi + yi * stride, stride, dx, dy);
+}
+
+/** Return the pixel at (x,y) using bilinear interpolation by clamping when out of borders. The image must be single channel input
+ *
+ * @warning Only works if the iterator was created with an IImage
+ *
+ * @param[in] first_pixel_ptr Pointer to the first pixel of a single channel image.
+ * @param[in] stride          Stride in bytes of the image
+ * @param[in] width           Width of the image
+ * @param[in] height          Height of the image
+ * @param[in] x               X position of the wanted pixel
+ * @param[in] y               Y position of the wanted pixel
+ *
+ * @return The pixel at (x, y) using bilinear interpolation.
+ */
+template <typename T>
+inline uint8_t
+pixel_bilinear_c1_clamp(const T *first_pixel_ptr, size_t stride, size_t width, size_t height, float x, float y)
+{
+    ARM_COMPUTE_ERROR_ON(first_pixel_ptr == nullptr);
+
+    x = std::max(-1.f, std::min(x, static_cast<float>(width)));
+    y = std::max(-1.f, std::min(y, static_cast<float>(height)));
+
+    const float xi = std::floor(x);
+    const float yi = std::floor(y);
+
+    const float dx = x - xi;
+    const float dy = y - yi;
+
+    if(dx == 0.0f)
+    {
+        if(dy == 0.0f)
+        {
+            return static_cast<T>(first_pixel_ptr[static_cast<int32_t>(xi) + static_cast<int32_t>(yi) * stride]);
+        }
+        return delta_linear_c1_y(first_pixel_ptr + static_cast<int32_t>(xi) + static_cast<int32_t>(yi) * stride,
+                                 stride, dy);
+    }
+    if(dy == 0.0f)
+    {
+        return delta_linear_c1_x(first_pixel_ptr + static_cast<int32_t>(xi) + static_cast<int32_t>(yi) * stride,
+                                 dx);
+    }
+    return delta_bilinear_c1(first_pixel_ptr + static_cast<int32_t>(xi) + static_cast<int32_t>(yi) * stride, stride,
+                             dx, dy);
+}
+
+/** Return the pixel at (x,y) using area interpolation by clamping when out of borders. The image must be single channel U8
+ *
+ * @note The interpolation area depends on the width and height ration of the input and output images
+ * @note Currently average of the contributing pixels is calculated
+ *
+ * @param[in] first_pixel_ptr Pointer to the first pixel of a single channel U8 image.
+ * @param[in] stride          Stride in bytes of the image
+ * @param[in] width           Width of the image
+ * @param[in] height          Height of the image
+ * @param[in] wr              Width ratio among the input image width and output image width.
+ * @param[in] hr              Height ratio among the input image height and output image height.
+ * @param[in] x               X position of the wanted pixel
+ * @param[in] y               Y position of the wanted pixel
+ *
+ * @return The pixel at (x, y) using area interpolation.
+ */
+inline uint8_t
+pixel_area_c1u8_clamp(const uint8_t *first_pixel_ptr, size_t stride, size_t width, size_t height, float wr,
+                      float hr, int x, int y)
+{
+    ARM_COMPUTE_ERROR_ON(first_pixel_ptr == nullptr);
+
+    // Calculate sampling position
+    float in_x = (x + 0.5f) * wr - 0.5f;
+    float in_y = (y + 0.5f) * hr - 0.5f;
+
+    // Get bounding box offsets
+    int x_from = std::floor(x * wr - 0.5f - in_x);
+    int y_from = std::floor(y * hr - 0.5f - in_y);
+    int x_to   = std::ceil((x + 1) * wr - 0.5f - in_x);
+    int y_to   = std::ceil((y + 1) * hr - 0.5f - in_y);
+
+    // Clamp position to borders
+    in_x = std::max(-1.f, std::min(in_x, static_cast<float>(width)));
+    in_y = std::max(-1.f, std::min(in_y, static_cast<float>(height)));
+
+    // Clamp bounding box offsets to borders
+    x_from = ((in_x + x_from) < -1) ? -1 : x_from;
+    y_from = ((in_y + y_from) < -1) ? -1 : y_from;
+    x_to   = ((in_x + x_to) > width) ? (width - in_x) : x_to;
+    y_to   = ((in_y + y_to) > height) ? (height - in_y) : y_to;
+
+    // Get pixel index
+    const int xi = std::floor(in_x);
+    const int yi = std::floor(in_y);
+
+    // Bounding box elements in each dimension
+    const int x_elements = (x_to - x_from + 1);
+    const int y_elements = (y_to - y_from + 1);
+    ARM_COMPUTE_ERROR_ON(x_elements == 0 || y_elements == 0);
+
+    // Sum pixels in area
+    int sum = 0;
+    for(int j = yi + y_from, je = yi + y_to; j <= je; ++j)
+    {
+        const uint8_t *ptr = first_pixel_ptr + j * stride + xi + x_from;
+        sum                = std::accumulate(ptr, ptr + x_elements, sum);
+    }
+
+    // Return average
+    return sum / (x_elements * y_elements);
+}
+} // namespace scale_helpers
+} // namespace arm_compute
+
+#endif /* SRC_CORE_HELPERS_SCALEHELPERS_H */
diff --git a/arm_compute/core/utils/misc/ICloneable.h b/src/core/helpers/SoftmaxHelpers.cpp
similarity index 68%
copy from arm_compute/core/utils/misc/ICloneable.h
copy to src/core/helpers/SoftmaxHelpers.cpp
index cbb0b3c..71b971a 100644
--- a/arm_compute/core/utils/misc/ICloneable.h
+++ b/src/core/helpers/SoftmaxHelpers.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+* Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,28 +21,25 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_MISC_ICLONEABLE_H
-#define ARM_COMPUTE_MISC_ICLONEABLE_H
-
-#include <memory>
+#include "src/core/helpers/SoftmaxHelpers.h"
 
 namespace arm_compute
 {
-namespace misc
+namespace softmax_helpers
 {
-/** Clonable Interface */
-template <class T>
-class ICloneable
+PermutationVector get_permutation_vector_from_softmax_axis(size_t axis)
 {
-public:
-    /** Default virtual desctructor */
-    virtual ~ICloneable() = default;
-    /** Provide a clone of the current object of class T
-     *
-     * @return Clone object of class T
-     */
-    virtual std::unique_ptr<T> clone() const = 0;
-};
-} // namespace misc
+    switch(axis)
+    {
+        case 1:
+            return PermutationVector(1U, 0U, 2U, 3U);
+        case 2:
+            return PermutationVector(2U, 1U, 0U, 3U);
+        case 3:
+            return PermutationVector(3U, 1U, 2U, 0U);
+        default:
+            ARM_COMPUTE_ERROR("Axis not supported");
+    }
+}
+} // namespace softmax_helpers
 } // namespace arm_compute
-#endif /* ARM_COMPUTE_MISC_ICLONEABLE_H */
diff --git a/src/core/helpers/SoftmaxHelpers.h b/src/core/helpers/SoftmaxHelpers.h
new file mode 100644
index 0000000..de5490a
--- /dev/null
+++ b/src/core/helpers/SoftmaxHelpers.h
@@ -0,0 +1,50 @@
+/*
+* Copyright (c) 2020 Arm Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef SRC_CORE_HELPERS_SOFTMAXHELPERS_H
+#define SRC_CORE_HELPERS_SOFTMAXHELPERS_H
+
+#include "arm_compute/core/Types.h"
+
+namespace arm_compute
+{
+namespace softmax_helpers
+{
+/** Given a softmax axis, this function returns the permutation vector required to put the axis to the front
+ *
+ * @note This function assumes a tensor rank <= 4
+ *
+ * Axis selects the dimension on which softmax is performed.
+ * E.g. For input of shape 4x5x6 and axis=1, softmax will be applied to 4x6=24 vectors of size 5.
+ * Interally softmax kernels is always performed on the first dimension (front dimension), therefore permutation is
+ * required to put the dimension specified by @p axis to the first dimension.
+ *
+ * @param[in] axis Axis on which to perform softmax. Supported: 1, 2, 3 (0 implies no permutation needed)
+ *
+ * @return the permutation vector
+ */
+PermutationVector get_permutation_vector_from_softmax_axis(size_t axis);
+} // namespace softmax_helpers
+} // namespace arm_compute
+
+#endif /* SRC_CORE_HELPERS_SOFTMAXHELPERS_H */
diff --git a/src/core/helpers/Utils.h b/src/core/helpers/Utils.h
new file mode 100644
index 0000000..3c3b2b9
--- /dev/null
+++ b/src/core/helpers/Utils.h
@@ -0,0 +1,97 @@
+/*
+* Copyright (c) 2020 Arm Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef SRC_CORE_HELPERS_UTILS_H
+#define SRC_CORE_HELPERS_UTILS_H
+
+#include "arm_compute/core/ITensorInfo.h"
+
+namespace arm_compute
+{
+/** Create a strides object based on the provided strides and the tensor dimensions.
+ *
+ * @param[in] info          Tensor info object providing the shape of the tensor for unspecified strides.
+ * @param[in] stride_x      Stride to be used in X dimension (in bytes).
+ * @param[in] fixed_strides Strides to be used in higher dimensions starting at Y (in bytes).
+ *
+ * @return Strides object based on the specified strides. Missing strides are
+ *         calculated based on the tensor shape and the strides of lower dimensions.
+ */
+template <typename T, typename... Ts>
+inline Strides compute_strides(const ITensorInfo &info, T stride_x, Ts &&... fixed_strides)
+{
+    const TensorShape &shape = info.tensor_shape();
+
+    // Create strides object
+    Strides strides(stride_x, fixed_strides...);
+
+    for(size_t i = 1 + sizeof...(Ts); i < info.num_dimensions(); ++i)
+    {
+        strides.set(i, shape[i - 1] * strides[i - 1]);
+    }
+
+    return strides;
+}
+
+/** Create a strides object based on the tensor dimensions.
+ *
+ * @param[in] info Tensor info object used to compute the strides.
+ *
+ * @return Strides object based on element size and tensor shape.
+ */
+template <typename... Ts>
+inline Strides compute_strides(const ITensorInfo &info)
+{
+    return compute_strides(info, info.element_size());
+}
+
+/** Given an integer value, this function returns the next power of two
+ *
+ * @param[in] x Input value
+ *
+ * @return the next power of two
+ */
+inline unsigned int get_next_power_two(unsigned int x)
+{
+    // Decrement by 1
+    x--;
+
+    // Shift right by 1
+    x |= x >> 1u;
+    // Shift right by 2
+    x |= x >> 2u;
+    // Shift right by 4
+    x |= x >> 4u;
+    // Shift right by 8
+    x |= x >> 8u;
+    // Shift right by 16
+    x |= x >> 16u;
+
+    // Increment by 1
+    x++;
+
+    return x;
+}
+} // namespace arm_compute
+
+#endif /* SRC_CORE_HELPERS_UTILS_H */
diff --git a/src/core/helpers/WindowHelpers.cpp b/src/core/helpers/WindowHelpers.cpp
new file mode 100644
index 0000000..ba10eb9
--- /dev/null
+++ b/src/core/helpers/WindowHelpers.cpp
@@ -0,0 +1,183 @@
+/*
+* Copyright (c) 2020 Arm Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include "src/core/helpers/WindowHelpers.h"
+
+namespace arm_compute
+{
+Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
+{
+    if(!skip_border)
+    {
+        border_size = BorderSize(0);
+    }
+
+    const Coordinates &anchor = valid_region.anchor;
+    const TensorShape &shape  = valid_region.shape;
+
+    Window window;
+
+    window.set(0, Window::Dimension(
+                   // Skip the border left of the image
+                   anchor[0] + border_size.left,
+                   // Skip the border right of the image
+                   // Make sure the window width is a multiple of the step size
+                   anchor[0] + border_size.left + ceil_to_multiple(std::max(0, static_cast<int>(shape[0]) - static_cast<int>(border_size.left) - static_cast<int>(border_size.right)), steps[0]),
+                   steps[0]));
+
+    size_t n = 1;
+
+    if(anchor.num_dimensions() > 1)
+    {
+        window.set(1, Window::Dimension(
+                       // Skip the border above the image
+                       anchor[1] + border_size.top,
+                       // Skip the border below the image
+                       anchor[1] + border_size.top + ceil_to_multiple(std::max(0, static_cast<int>(shape[1]) - static_cast<int>(border_size.top) - static_cast<int>(border_size.bottom)), steps[1]),
+                       steps[1]));
+
+        ++n;
+    }
+
+    if(anchor.num_dimensions() > 2)
+    {
+        window.set(2, Window::Dimension(anchor[2], std::max<size_t>(1, shape[2]), steps[2]));
+
+        ++n;
+    }
+
+    for(; n < anchor.num_dimensions(); ++n)
+    {
+        window.set(n, Window::Dimension(anchor[n], std::max<size_t>(1, shape[n])));
+    }
+
+    for(; n < Coordinates::num_max_dimensions; ++n)
+    {
+        window.set(n, Window::Dimension(0, 1));
+    }
+
+    return window;
+}
+
+Window calculate_max_enlarged_window(const ValidRegion &valid_region, const Steps &steps, BorderSize border_size)
+{
+    const Coordinates &anchor = valid_region.anchor;
+    const TensorShape &shape  = valid_region.shape;
+
+    Window window;
+
+    window.set(0, Window::Dimension(
+                   // move the anchor to the start from the border
+                   anchor[0] - border_size.left,
+                   // move the anchor to include the right end border
+                   // Make sure the window width is a multiple of the step size
+                   anchor[0] - border_size.left + ceil_to_multiple(shape[0] + border_size.left + border_size.right, steps[0]),
+                   steps[0]));
+
+    size_t n = 1;
+
+    if(anchor.num_dimensions() > 1)
+    {
+        window.set(1, Window::Dimension(
+                       // Include the border above the image
+                       anchor[1] - border_size.top,
+                       // Include the border below the image
+                       anchor[1] - border_size.top + ceil_to_multiple(shape[1] + border_size.top + border_size.bottom, steps[1]),
+                       steps[1]));
+
+        ++n;
+    }
+
+    if(anchor.num_dimensions() > 2)
+    {
+        window.set(2, Window::Dimension(0, std::max<size_t>(1, shape[n]), steps[2]));
+
+        ++n;
+    }
+
+    for(; n < anchor.num_dimensions(); ++n)
+    {
+        window.set(n, Window::Dimension(anchor[n], std::max<size_t>(1, shape[n])));
+    }
+
+    for(; n < Coordinates::num_max_dimensions; ++n)
+    {
+        window.set(n, Window::Dimension(0, 1));
+    }
+
+    return window;
+}
+
+Window calculate_max_window_horizontal(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
+{
+    if(skip_border)
+    {
+        border_size.top    = 0;
+        border_size.bottom = 0;
+    }
+    else
+    {
+        border_size.left  = 0;
+        border_size.right = 0;
+    }
+
+    const Coordinates &anchor = valid_region.anchor;
+    const TensorShape &shape  = valid_region.shape;
+
+    Window window;
+
+    window.set(0, Window::Dimension(
+                   // Skip the border left of the image
+                   anchor[0] + border_size.left,
+                   // Skip the border right of the image
+                   // Make sure the window width is a multiple of the step size
+                   anchor[0] + border_size.left + ceil_to_multiple(std::max(0, static_cast<int>(shape[0]) - static_cast<int>(border_size.left) - static_cast<int>(border_size.right)), steps[0]),
+                   steps[0]));
+
+    size_t n = 1;
+
+    if(anchor.num_dimensions() > 1)
+    {
+        window.set(1, Window::Dimension(
+                       // Skip the border above the image
+                       anchor[1] - border_size.top,
+                       // Skip the border below the image
+                       anchor[1] + shape[1] + border_size.bottom,
+                       1));
+
+        ++n;
+    }
+
+    for(; n < anchor.num_dimensions(); ++n)
+    {
+        window.set(n, Window::Dimension(anchor[n], std::max<size_t>(1, shape[n])));
+    }
+
+    for(; n < Coordinates::num_max_dimensions; ++n)
+    {
+        window.set(n, Window::Dimension(0, 1));
+    }
+
+    return window;
+}
+} // namespace arm_compute
diff --git a/src/core/helpers/WindowHelpers.h b/src/core/helpers/WindowHelpers.h
new file mode 100644
index 0000000..9bc2135
--- /dev/null
+++ b/src/core/helpers/WindowHelpers.h
@@ -0,0 +1,172 @@
+/*
+* Copyright (c) 2020 Arm Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef SRC_CORE_HELPERS_WINDOWHELPERS_H
+#define SRC_CORE_HELPERS_WINDOWHELPERS_H
+
+#include "arm_compute/core/IAccessWindow.h"
+#include "arm_compute/core/Steps.h"
+#include "arm_compute/core/Window.h"
+
+namespace arm_compute
+{
+/** Update window and padding size for each of the access patterns.
+ *
+ * First the window size is reduced based on all access patterns that are not
+ * allowed to modify the padding of the underlying tensor. Then the padding of
+ * the remaining tensors is increased to match the window.
+ *
+ * @param[in] win      Window that is used by the kernel.
+ * @param[in] patterns Access patterns used to calculate the final window and padding.
+ *
+ * @return True if the window has been changed. Changes to the padding do not
+ *         influence the returned value.
+ */
+template <typename... Ts>
+bool update_window_and_padding(Window &win, Ts &&... patterns)
+{
+    bool window_changed = false;
+
+    utility::for_each([&](const IAccessWindow & w)
+    {
+        window_changed |= w.update_window_if_needed(win);
+    },
+    patterns...);
+
+    bool padding_changed = false;
+
+    utility::for_each([&](IAccessWindow & w)
+    {
+        padding_changed |= w.update_padding_if_needed(win);
+    },
+    patterns...);
+
+    return window_changed;
+}
+
+/** Intersect multiple valid regions.
+ *
+ * @param[in] regions Valid regions.
+ *
+ * @return Intersection of all regions.
+ */
+template <typename... Ts>
+ValidRegion intersect_valid_regions(const Ts &... regions)
+{
+    auto intersect = [](const ValidRegion & r1, const ValidRegion & r2) -> ValidRegion
+    {
+        ValidRegion region;
+
+        for(size_t d = 0; d < std::min(r1.anchor.num_dimensions(), r2.anchor.num_dimensions()); ++d)
+        {
+            region.anchor.set(d, std::max(r1.anchor[d], r2.anchor[d]));
+        }
+
+        for(size_t d = 0; d < std::min(r1.shape.num_dimensions(), r2.shape.num_dimensions()); ++d)
+        {
+            region.shape.set(d, std::min(r1.shape[d], r2.shape[d]));
+        }
+
+        return region;
+    };
+
+    return utility::foldl(intersect, regions...);
+}
+
+#ifndef DOXYGEN_SKIP_THIS
+/** Calculate the maximum window for a given tensor shape and border setting
+ *
+ * @param[in] valid_region Valid region object defining the shape of the tensor space for which the window is created.
+ * @param[in] steps        (Optional) Number of elements processed for each step.
+ * @param[in] skip_border  (Optional) If true exclude the border region from the window.
+ * @param[in] border_size  (Optional) Border size.
+ *
+ * @return The maximum window the kernel can be executed on.
+ */
+Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps = Steps(), bool skip_border = false, BorderSize border_size = BorderSize());
+
+/** Calculate the maximum window for a given tensor shape and border setting
+ *
+ * @param[in] info        Tensor info object defining the shape of the object for which the window is created.
+ * @param[in] steps       (Optional) Number of elements processed for each step.
+ * @param[in] skip_border (Optional) If true exclude the border region from the window.
+ * @param[in] border_size (Optional) Border size.
+ *
+ * @return The maximum window the kernel can be executed on.
+ */
+inline Window calculate_max_window(const ITensorInfo &info, const Steps &steps = Steps(), bool skip_border = false, BorderSize border_size = BorderSize())
+{
+    return calculate_max_window(info.valid_region(), steps, skip_border, border_size);
+}
+
+/** Calculate the maximum window used by a horizontal kernel for a given tensor shape and border setting
+ *
+ * @param[in] valid_region Valid region object defining the shape of the tensor space for which the window is created.
+ * @param[in] steps        (Optional) Number of elements processed for each step.
+ * @param[in] skip_border  (Optional) If true exclude the border region from the window.
+ * @param[in] border_size  (Optional) Border size. The border region will be excluded from the window.
+ *
+ * @return The maximum window the kernel can be executed on.
+ */
+Window calculate_max_window_horizontal(const ValidRegion &valid_region, const Steps &steps = Steps(), bool skip_border = false, BorderSize border_size = BorderSize());
+
+/** Calculate the maximum window used by a horizontal kernel for a given tensor shape and border setting
+ *
+ * @param[in] info        Tensor info object defining the shape of the object for which the window is created.
+ * @param[in] steps       (Optional) Number of elements processed for each step.
+ * @param[in] skip_border (Optional) If true exclude the border region from the window.
+ * @param[in] border_size (Optional) Border size.
+ *
+ * @return The maximum window the kernel can be executed on.
+ */
+inline Window calculate_max_window_horizontal(const ITensorInfo &info, const Steps &steps = Steps(), bool skip_border = false, BorderSize border_size = BorderSize())
+{
+    return calculate_max_window_horizontal(info.valid_region(), steps, skip_border, border_size);
+}
+
+/** Calculate the maximum window for a given tensor shape and border setting. The window will also includes the border.
+ *
+ * @param[in] valid_region Valid region object defining the shape of the tensor space for which the window is created.
+ * @param[in] steps        (Optional) Number of elements processed for each step.
+ * @param[in] border_size  (Optional) Border size. The border region will be included in the window.
+ *
+ * @return The maximum window the kernel can be executed on.
+ */
+Window calculate_max_enlarged_window(const ValidRegion &valid_region, const Steps &steps = Steps(), BorderSize border_size = BorderSize());
+
+/** Calculate the maximum window for a given tensor shape and border setting. The window will also includes the border.
+ *
+ * @param[in] info        Tensor info object defining the shape of the object for which the window is created.
+ * @param[in] steps       (Optional) Number of elements processed for each step.
+ * @param[in] border_size (Optional) Border size. The border region will be included in the window.
+ *
+ * @return The maximum window the kernel can be executed on.
+ */
+inline Window calculate_max_enlarged_window(const ITensorInfo &info, const Steps &steps = Steps(), BorderSize border_size = BorderSize())
+{
+    return calculate_max_enlarged_window(info.valid_region(), steps, border_size);
+}
+#endif /* DOXYGEN_SKIP_THIS */
+} // namespace arm_compute
+
+#endif /* SRC_CORE_HELPERS_WINDOWHELPERS_H */
diff --git a/arm_compute/core/utils/helpers/bit_ops.h b/src/core/utils/helpers/bit_ops.h
similarity index 94%
rename from arm_compute/core/utils/helpers/bit_ops.h
rename to src/core/utils/helpers/bit_ops.h
index eee360c..ef60214 100644
--- a/arm_compute/core/utils/helpers/bit_ops.h
+++ b/src/core/utils/helpers/bit_ops.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -24,7 +24,7 @@
 #ifndef ARM_COMPUTE_UTILS_HELPERS_BIT_OPS_H
 #define ARM_COMPUTE_UTILS_HELPERS_BIT_OPS_H
 
-#include "arm_compute/core/utils/misc/Requires.h"
+#include "support/Requires.h"
 
 #include <type_traits>
 
diff --git a/src/core/utils/helpers/fft.cpp b/src/core/utils/helpers/fft.cpp
index 4c2f8fa..64633c6 100644
--- a/src/core/utils/helpers/fft.cpp
+++ b/src/core/utils/helpers/fft.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/core/utils/helpers/fft.h"
+#include "src/core/utils/helpers/fft.h"
 
 #include <numeric>
 
diff --git a/arm_compute/core/utils/helpers/fft.h b/src/core/utils/helpers/fft.h
similarity index 97%
rename from arm_compute/core/utils/helpers/fft.h
rename to src/core/utils/helpers/fft.h
index 7d111b7..f7b99dd 100644
--- a/arm_compute/core/utils/helpers/fft.h
+++ b/src/core/utils/helpers/fft.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/arm_compute/core/utils/helpers/float_ops.h b/src/core/utils/helpers/float_ops.h
similarity index 98%
rename from arm_compute/core/utils/helpers/float_ops.h
rename to src/core/utils/helpers/float_ops.h
index 1a08fc7..a475a23 100644
--- a/arm_compute/core/utils/helpers/float_ops.h
+++ b/src/core/utils/helpers/float_ops.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/arm_compute/core/utils/helpers/tensor_info.h b/src/core/utils/helpers/tensor_info.h
similarity index 98%
rename from arm_compute/core/utils/helpers/tensor_info.h
rename to src/core/utils/helpers/tensor_info.h
index 4432340..9279532 100644
--- a/arm_compute/core/utils/helpers/tensor_info.h
+++ b/src/core/utils/helpers/tensor_info.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/src/core/utils/helpers/tensor_transform.cpp b/src/core/utils/helpers/tensor_transform.cpp
index 84302ea..f221699 100644
--- a/src/core/utils/helpers/tensor_transform.cpp
+++ b/src/core/utils/helpers/tensor_transform.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,7 +23,7 @@
  */
 #include "arm_compute/core/utils/helpers/tensor_transform.h"
 
-#include "arm_compute/core/utils/helpers/bit_ops.h"
+#include "bit_ops.h"
 
 namespace arm_compute
 {
diff --git a/src/graph/algorithms/TopologicalSort.cpp b/src/graph/algorithms/TopologicalSort.cpp
index 3647e13..3a69352 100644
--- a/src/graph/algorithms/TopologicalSort.cpp
+++ b/src/graph/algorithms/TopologicalSort.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -25,7 +25,7 @@
 
 #include "arm_compute/graph/Graph.h"
 
-#include "arm_compute/core/utils/misc/Iterable.h"
+#include "support/Iterable.h"
 
 #include <list>
 #include <stack>
@@ -185,4 +185,4 @@
     return dfs_order_vector;
 }
 } // namespace graph
-} // namespace arm_compute
\ No newline at end of file
+} // namespace arm_compute
diff --git a/src/graph/backends/CL/CLFunctionsFactory.cpp b/src/graph/backends/CL/CLFunctionsFactory.cpp
index 5f2f46f..cd73255 100644
--- a/src/graph/backends/CL/CLFunctionsFactory.cpp
+++ b/src/graph/backends/CL/CLFunctionsFactory.cpp
@@ -23,12 +23,12 @@
  */
 #include "arm_compute/graph/backends/CL/CLFunctionFactory.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/graph/Graph.h"
 #include "arm_compute/graph/GraphContext.h"
 #include "arm_compute/graph/backends/FunctionHelpers.h"
 #include "arm_compute/runtime/CL/CLFunctions.h"
 #include "arm_compute/runtime/CPP/CPPFunctions.h"
+#include "support/Cast.h"
 
 using namespace arm_compute::utils::cast;
 
diff --git a/src/graph/backends/CL/CLNodeValidator.cpp b/src/graph/backends/CL/CLNodeValidator.cpp
index 8b2ecaf..8c1fedd 100644
--- a/src/graph/backends/CL/CLNodeValidator.cpp
+++ b/src/graph/backends/CL/CLNodeValidator.cpp
@@ -26,9 +26,9 @@
 #include "arm_compute/graph/backends/ValidateHelpers.h"
 #include "arm_compute/graph/nodes/Nodes.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/runtime/CL/CLFunctions.h"
 #include "arm_compute/runtime/CPP/CPPFunctions.h"
+#include "support/Cast.h"
 
 using namespace arm_compute::utils::cast;
 
diff --git a/src/graph/backends/CL/CLSubTensorHandle.cpp b/src/graph/backends/CL/CLSubTensorHandle.cpp
index ada0d68..b97d258 100644
--- a/src/graph/backends/CL/CLSubTensorHandle.cpp
+++ b/src/graph/backends/CL/CLSubTensorHandle.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,7 +23,7 @@
  */
 #include "arm_compute/graph/backends/CL/CLSubTensorHandle.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "support/Cast.h"
 
 namespace arm_compute
 {
diff --git a/src/graph/backends/GLES/GCFunctionsFactory.cpp b/src/graph/backends/GLES/GCFunctionsFactory.cpp
index 8ecb593..7d9d388 100644
--- a/src/graph/backends/GLES/GCFunctionsFactory.cpp
+++ b/src/graph/backends/GLES/GCFunctionsFactory.cpp
@@ -23,11 +23,11 @@
  */
 #include "arm_compute/graph/backends/GLES/GCFunctionFactory.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/graph/Graph.h"
 #include "arm_compute/graph/GraphContext.h"
 #include "arm_compute/graph/backends/FunctionHelpers.h"
 #include "arm_compute/runtime/GLES_COMPUTE/GCFunctions.h"
+#include "support/Cast.h"
 
 using namespace arm_compute::utils::cast;
 
diff --git a/src/graph/backends/GLES/GCNodeValidator.cpp b/src/graph/backends/GLES/GCNodeValidator.cpp
index 159e512..13a93a2 100644
--- a/src/graph/backends/GLES/GCNodeValidator.cpp
+++ b/src/graph/backends/GLES/GCNodeValidator.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -26,8 +26,8 @@
 #include "arm_compute/graph/backends/ValidateHelpers.h"
 #include "arm_compute/graph/nodes/Nodes.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/runtime/GLES_COMPUTE/GCFunctions.h"
+#include "support/Cast.h"
 
 using namespace arm_compute::utils::cast;
 
diff --git a/src/graph/backends/NEON/NEFunctionFactory.cpp b/src/graph/backends/NEON/NEFunctionFactory.cpp
index 61df45d..95c6631 100644
--- a/src/graph/backends/NEON/NEFunctionFactory.cpp
+++ b/src/graph/backends/NEON/NEFunctionFactory.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/graph/backends/NEON/NEFunctionFactory.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/graph/Graph.h"
 #include "arm_compute/graph/GraphContext.h"
 #include "arm_compute/graph/Logger.h"
@@ -33,6 +32,7 @@
 #include "arm_compute/graph/nodes/Nodes.h"
 #include "arm_compute/runtime/CPP/CPPFunctions.h"
 #include "arm_compute/runtime/NEON/NEFunctions.h"
+#include "support/Cast.h"
 #include "support/ToolchainSupport.h"
 
 using namespace arm_compute::utils::cast;
diff --git a/src/graph/backends/NEON/NENodeValidator.cpp b/src/graph/backends/NEON/NENodeValidator.cpp
index 19c96ea..63e8ff9 100644
--- a/src/graph/backends/NEON/NENodeValidator.cpp
+++ b/src/graph/backends/NEON/NENodeValidator.cpp
@@ -26,9 +26,9 @@
 #include "arm_compute/graph/backends/ValidateHelpers.h"
 #include "arm_compute/graph/nodes/Nodes.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/runtime/CPP/CPPFunctions.h"
 #include "arm_compute/runtime/NEON/NEFunctions.h"
+#include "support/Cast.h"
 
 using namespace arm_compute::utils::cast;
 
diff --git a/src/graph/backends/NEON/NETensorHandle.cpp b/src/graph/backends/NEON/NETensorHandle.cpp
index c8fc3f1..4393156 100644
--- a/src/graph/backends/NEON/NETensorHandle.cpp
+++ b/src/graph/backends/NEON/NETensorHandle.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,8 +23,8 @@
  */
 #include "arm_compute/graph/backends/NEON/NETensorHandle.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/runtime/MemoryGroup.h"
+#include "support/Cast.h"
 
 namespace arm_compute
 {
diff --git a/src/graph/detail/CrossLayerMemoryManagerHelpers.cpp b/src/graph/detail/CrossLayerMemoryManagerHelpers.cpp
index fd16625..b45f453 100644
--- a/src/graph/detail/CrossLayerMemoryManagerHelpers.cpp
+++ b/src/graph/detail/CrossLayerMemoryManagerHelpers.cpp
@@ -33,7 +33,7 @@
 #include "arm_compute/graph/backends/BackendRegistry.h"
 
 #include "arm_compute/core/ITensor.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "support/Cast.h"
 
 #include <algorithm>
 #include <map>
diff --git a/src/graph/mutators/DepthConcatSubTensorMutator.cpp b/src/graph/mutators/DepthConcatSubTensorMutator.cpp
index fa63f56..963b948 100644
--- a/src/graph/mutators/DepthConcatSubTensorMutator.cpp
+++ b/src/graph/mutators/DepthConcatSubTensorMutator.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,8 +30,8 @@
 #include "arm_compute/graph/backends/BackendRegistry.h"
 #include "arm_compute/graph/nodes/ConcatenateLayerNode.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
-#include "arm_compute/core/utils/misc/Iterable.h"
+#include "support/Cast.h"
+#include "support/Iterable.h"
 
 namespace arm_compute
 {
diff --git a/src/graph/mutators/GroupedConvolutionMutator.cpp b/src/graph/mutators/GroupedConvolutionMutator.cpp
index e3d3812..b7c551c 100644
--- a/src/graph/mutators/GroupedConvolutionMutator.cpp
+++ b/src/graph/mutators/GroupedConvolutionMutator.cpp
@@ -30,7 +30,7 @@
 #include "arm_compute/graph/backends/BackendRegistry.h"
 #include "arm_compute/graph/nodes/Nodes.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "support/Cast.h"
 
 #include "support/StringSupport.h"
 
diff --git a/src/graph/mutators/NodeExecutionMethodMutator.cpp b/src/graph/mutators/NodeExecutionMethodMutator.cpp
index 48bb9f7..09a3cf5 100644
--- a/src/graph/mutators/NodeExecutionMethodMutator.cpp
+++ b/src/graph/mutators/NodeExecutionMethodMutator.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,7 +29,7 @@
 #include "arm_compute/graph/backends/BackendRegistry.h"
 #include "arm_compute/graph/nodes/Nodes.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "support/Cast.h"
 
 namespace arm_compute
 {
diff --git a/src/graph/mutators/NodeFusionMutator.cpp b/src/graph/mutators/NodeFusionMutator.cpp
index 2a80825..1d47668 100644
--- a/src/graph/mutators/NodeFusionMutator.cpp
+++ b/src/graph/mutators/NodeFusionMutator.cpp
@@ -30,7 +30,7 @@
 #include "arm_compute/graph/nodes/FusedConvolutionBatchNormalizationNode.h"
 #include "arm_compute/graph/nodes/Nodes.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "support/Cast.h"
 
 #include <set>
 
@@ -300,11 +300,12 @@
 void NodeFusionMutator::mutate(Graph &g)
 {
     // Supported activations when fusing
-    const std::set<Activation> supported_fused_activations    = { Activation::ABS, Activation::BOUNDED_RELU, Activation::ELU,
-                                                                  Activation::HARD_SWISH, Activation::IDENTITY, Activation::LEAKY_RELU,
-                                                                  Activation::LINEAR, Activation::LOGISTIC, Activation::LU_BOUNDED_RELU,
-                                                                  Activation::RELU, Activation::SOFT_RELU, Activation::SQRT,
-                                                                  Activation::SQUARE, Activation::TANH  };
+    const std::set<Activation> supported_fused_activations = { Activation::ABS, Activation::BOUNDED_RELU, Activation::ELU,
+                                                               Activation::HARD_SWISH, Activation::IDENTITY, Activation::LEAKY_RELU,
+                                                               Activation::LINEAR, Activation::LOGISTIC, Activation::LU_BOUNDED_RELU,
+                                                               Activation::RELU, Activation::SOFT_RELU, Activation::SQRT,
+                                                               Activation::SQUARE, Activation::TANH
+                                                             };
 
     // Preconditions
     auto empty_prec = [](INode &)
diff --git a/src/graph/mutators/SplitLayerSubTensorMutator.cpp b/src/graph/mutators/SplitLayerSubTensorMutator.cpp
index 359bba4..2c28a1a 100644
--- a/src/graph/mutators/SplitLayerSubTensorMutator.cpp
+++ b/src/graph/mutators/SplitLayerSubTensorMutator.cpp
@@ -30,8 +30,8 @@
 #include "arm_compute/graph/backends/BackendRegistry.h"
 #include "arm_compute/graph/nodes/SplitLayerNode.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
-#include "arm_compute/core/utils/misc/Iterable.h"
+#include "support/Cast.h"
+#include "support/Iterable.h"
 
 namespace arm_compute
 {
diff --git a/src/graph/mutators/SyntheticDataTypeMutator.cpp b/src/graph/mutators/SyntheticDataTypeMutator.cpp
index dbbebdf..532c0e8 100644
--- a/src/graph/mutators/SyntheticDataTypeMutator.cpp
+++ b/src/graph/mutators/SyntheticDataTypeMutator.cpp
@@ -29,7 +29,7 @@
 #include "arm_compute/graph/Utils.h"
 #include "arm_compute/graph/nodes/Nodes.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "support/Cast.h"
 
 #include <set>
 
diff --git a/src/runtime/CL/CLHelpers.cpp b/src/runtime/CL/CLHelpers.cpp
index adfdc3c..5f1842f7 100644
--- a/src/runtime/CL/CLHelpers.cpp
+++ b/src/runtime/CL/CLHelpers.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -26,6 +26,7 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/Error.h"
+#include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/CL/CLRuntimeContext.h"
 
 namespace
diff --git a/src/runtime/CL/CLMemory.cpp b/src/runtime/CL/CLMemory.cpp
index efbc68f..a1743c5 100644
--- a/src/runtime/CL/CLMemory.cpp
+++ b/src/runtime/CL/CLMemory.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/CL/CLMemory.h"
 
 #include "arm_compute/core/Error.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "support/Cast.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/CLRuntimeContext.cpp b/src/runtime/CL/CLRuntimeContext.cpp
index 2fc7f93..571e309 100644
--- a/src/runtime/CL/CLRuntimeContext.cpp
+++ b/src/runtime/CL/CLRuntimeContext.cpp
@@ -26,6 +26,8 @@
 #include "arm_compute/runtime/CL/CLHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 
+#include "support/MemorySupport.h"
+
 namespace arm_compute
 {
 CLRuntimeContext::CLRuntimeContext()
diff --git a/src/runtime/CL/CLTensorAllocator.cpp b/src/runtime/CL/CLTensorAllocator.cpp
index 90d7788..f37fc77 100644
--- a/src/runtime/CL/CLTensorAllocator.cpp
+++ b/src/runtime/CL/CLTensorAllocator.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2019 Arm Limited.
+ * Copyright (c) 2016-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -28,6 +28,8 @@
 #include "arm_compute/runtime/CL/CLRuntimeContext.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 
+#include "support/MemorySupport.h"
+
 namespace arm_compute
 {
 const cl::Buffer CLTensorAllocator::_empty_buffer = cl::Buffer();
diff --git a/src/runtime/CL/functions/CLArgMinMaxLayer.cpp b/src/runtime/CL/functions/CLArgMinMaxLayer.cpp
index ad6e7ba..57c4f68 100644
--- a/src/runtime/CL/functions/CLArgMinMaxLayer.cpp
+++ b/src/runtime/CL/functions/CLArgMinMaxLayer.cpp
@@ -24,13 +24,14 @@
 
 #include "arm_compute/runtime/CL/functions/CLArgMinMaxLayer.h"
 
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
-#include "arm_compute/runtime/Utils.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/runtime/Utils.h"
 
 namespace arm_compute
 {
@@ -47,7 +48,7 @@
     ARM_COMPUTE_RETURN_ERROR_ON_MSG(op != ReductionOperation::ARG_IDX_MAX && op != ReductionOperation::ARG_IDX_MIN, "Invalid reduction operation");
     ARM_COMPUTE_RETURN_ERROR_ON_MSG(axis >= static_cast<int>(TensorShape::num_max_dimensions), "Reduction axis greater than max number of dimensions");
     ARM_COMPUTE_RETURN_ERROR_ON_MSG(axis > 3, "Unsupported reduction axis");
-    const unsigned int num_of_stages = calculate_number_of_stages_only_x_axis(input->dimension(0), axis);
+    const unsigned int num_of_stages = utils::calculate_number_of_stages_only_x_axis(input->dimension(0), axis);
 
     DataType   output_data_type = DataType::S32;
     TensorInfo not_reshaped_output;
@@ -115,7 +116,7 @@
 void CLArgMinMaxLayer::configure(const CLCompileContext &compile_context, const ICLTensor *input, int axis, ICLTensor *output, const ReductionOperation &op)
 {
     ARM_COMPUTE_ERROR_ON_NULLPTR(input, output);
-    _num_of_stages  = calculate_number_of_stages_only_x_axis(input->info()->dimension(0), axis);
+    _num_of_stages  = utils::calculate_number_of_stages_only_x_axis(input->info()->dimension(0), axis);
     _reduction_axis = axis;
 
     const TensorShape output_shape     = arm_compute::misc::shape_calculator::compute_reduced_shape(input->info()->tensor_shape(), axis, false);
@@ -172,4 +173,4 @@
     }
     _reshape.run();
 }
-} // namespace arm_compute
\ No newline at end of file
+} // namespace arm_compute
diff --git a/src/runtime/CL/functions/CLConcatenateLayer.cpp b/src/runtime/CL/functions/CLConcatenateLayer.cpp
index 4214813..2eb310b 100644
--- a/src/runtime/CL/functions/CLConcatenateLayer.cpp
+++ b/src/runtime/CL/functions/CLConcatenateLayer.cpp
@@ -36,6 +36,7 @@
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
+#include "src/core/helpers/AutoConfiguration.h"
 #include "support/MemorySupport.h"
 
 namespace arm_compute
diff --git a/src/runtime/CL/functions/CLConvertFullyConnectedWeights.cpp b/src/runtime/CL/functions/CLConvertFullyConnectedWeights.cpp
index 4c78767..b291ae5 100644
--- a/src/runtime/CL/functions/CLConvertFullyConnectedWeights.cpp
+++ b/src/runtime/CL/functions/CLConvertFullyConnectedWeights.cpp
@@ -23,6 +23,8 @@
  */
 #include "arm_compute/runtime/CL/functions/CLConvertFullyConnectedWeights.h"
 
+#include "support/MemorySupport.h"
+
 namespace arm_compute
 {
 void CLConvertFullyConnectedWeights::configure(const ICLTensor *input, ICLTensor *output, const TensorShape &original_input_shape,
diff --git a/src/runtime/CL/functions/CLConvolutionLayer.cpp b/src/runtime/CL/functions/CLConvolutionLayer.cpp
index 630352e..85355f0 100644
--- a/src/runtime/CL/functions/CLConvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLConvolutionLayer.cpp
@@ -30,6 +30,8 @@
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 
+#include "support/MemorySupport.h"
+
 #include <cmath>
 #include <memory>
 #include <tuple>
diff --git a/src/runtime/CL/functions/CLCropResize.cpp b/src/runtime/CL/functions/CLCropResize.cpp
index 529f7bf..6167e9d 100644
--- a/src/runtime/CL/functions/CLCropResize.cpp
+++ b/src/runtime/CL/functions/CLCropResize.cpp
@@ -25,6 +25,10 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/WindowHelpers.h"
+
+#include "support/MemorySupport.h"
 
 #include <cstddef>
 
diff --git a/src/runtime/CL/functions/CLDeconvolutionLayer.cpp b/src/runtime/CL/functions/CLDeconvolutionLayer.cpp
index cd55336..e6717b6 100644
--- a/src/runtime/CL/functions/CLDeconvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLDeconvolutionLayer.cpp
@@ -29,6 +29,8 @@
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 
+#include "support/MemorySupport.h"
+
 #include <cmath>
 #include <memory>
 #include <tuple>
diff --git a/src/runtime/CL/functions/CLDirectConvolutionLayer.cpp b/src/runtime/CL/functions/CLDirectConvolutionLayer.cpp
index c1055dd..07e7a18 100644
--- a/src/runtime/CL/functions/CLDirectConvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLDirectConvolutionLayer.cpp
@@ -43,7 +43,7 @@
 }
 
 void CLDirectConvolutionLayer::configure(const CLCompileContext &compile_context, ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output,
-                                         const PadStrideInfo &conv_info,
+                                         const PadStrideInfo       &conv_info,
                                          const ActivationLayerInfo &act_info)
 {
     // Set GPU target
diff --git a/src/runtime/CL/functions/CLDirectDeconvolutionLayer.cpp b/src/runtime/CL/functions/CLDirectDeconvolutionLayer.cpp
index 3515c25..0ffafa0 100644
--- a/src/runtime/CL/functions/CLDirectDeconvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLDirectDeconvolutionLayer.cpp
@@ -28,6 +28,7 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 #include <memory>
 #include <tuple>
diff --git a/src/runtime/CL/functions/CLFFT1D.cpp b/src/runtime/CL/functions/CLFFT1D.cpp
index 7d15d33..1269cba 100644
--- a/src/runtime/CL/functions/CLFFT1D.cpp
+++ b/src/runtime/CL/functions/CLFFT1D.cpp
@@ -25,8 +25,8 @@
 
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/utils/helpers/fft.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
+#include "src/core/utils/helpers/fft.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLFFTConvolutionLayer.cpp b/src/runtime/CL/functions/CLFFTConvolutionLayer.cpp
index 1def674..4d0eab8 100644
--- a/src/runtime/CL/functions/CLFFTConvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLFFTConvolutionLayer.cpp
@@ -26,10 +26,13 @@
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/utils/helpers/fft.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 #include "arm_compute/runtime/CPP/CPPScheduler.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/utils/helpers/fft.h"
+
+#include "support/MemorySupport.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLFill.cpp b/src/runtime/CL/functions/CLFill.cpp
index 6c0f178..a89383e 100644
--- a/src/runtime/CL/functions/CLFill.cpp
+++ b/src/runtime/CL/functions/CLFill.cpp
@@ -26,6 +26,8 @@
 #include "arm_compute/core/CL/kernels/CLMemsetKernel.h"
 #include "arm_compute/core/Types.h"
 
+#include "support/MemorySupport.h"
+
 #include <utility>
 
 namespace arm_compute
diff --git a/src/runtime/CL/functions/CLFullyConnectedLayer.cpp b/src/runtime/CL/functions/CLFullyConnectedLayer.cpp
index 4f365b6..75e87c3 100644
--- a/src/runtime/CL/functions/CLFullyConnectedLayer.cpp
+++ b/src/runtime/CL/functions/CLFullyConnectedLayer.cpp
@@ -25,10 +25,10 @@
 
 #include "arm_compute/core/Size2D.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
+#include "support/Cast.h"
 #include "support/MemorySupport.h"
 
 #include <algorithm>
diff --git a/src/runtime/CL/functions/CLGEMM.cpp b/src/runtime/CL/functions/CLGEMM.cpp
index d56b341..ccae671 100644
--- a/src/runtime/CL/functions/CLGEMM.cpp
+++ b/src/runtime/CL/functions/CLGEMM.cpp
@@ -23,10 +23,7 @@
  */
 #include "arm_compute/runtime/CL/functions/CLGEMM.h"
 
-#include "arm_compute/core/CL/ICLGEMMKernelConfiguration.h"
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfiguration.h"
-#include "arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfiguration.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/GPUTarget.h"
 #include "arm_compute/core/Helpers.h"
@@ -35,12 +32,18 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/utils/helpers/float_ops.h"
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
-#include "arm_compute/runtime/CL/gemm/CLGEMMKernelSelection.h"
 #include "arm_compute/runtime/ITensorAllocator.h"
+#include "src/core/CL/ICLGEMMKernelConfiguration.h"
+#include "src/core/CL/gemm/reshaped/CLGEMMReshapedKernelConfiguration.h"
+#include "src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfiguration.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/utils/helpers/float_ops.h"
+#include "src/runtime/CL/gemm/CLGEMMKernelSelection.h"
+#include "support/Cast.h"
+
+#include "support/MemorySupport.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLGEMMConvolutionLayer.cpp b/src/runtime/CL/functions/CLGEMMConvolutionLayer.cpp
index ee90b39..e871b39 100644
--- a/src/runtime/CL/functions/CLGEMMConvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLGEMMConvolutionLayer.cpp
@@ -27,10 +27,11 @@
 #include "arm_compute/core/Size2D.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "support/Cast.h"
 
 #include <cmath>
 #include <memory>
diff --git a/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp b/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp
index 30dce5b..7a8de6c 100644
--- a/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp
+++ b/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp
@@ -24,8 +24,6 @@
 #include "arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h"
 
 #include "arm_compute/core/CL/ICLTensor.h"
-#include "arm_compute/core/CL/gemm/native/CLGEMMNativeKernelConfiguration.h"
-#include "arm_compute/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfiguration.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/KernelDescriptors.h"
@@ -35,7 +33,10 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
-#include "arm_compute/runtime/CL/gemm/CLGEMMKernelSelection.h"
+#include "src/core/CL/gemm/native/CLGEMMNativeKernelConfiguration.h"
+#include "src/core/CL/gemm/reshaped_only_rhs/CLGEMMReshapedOnlyRHSKernelConfiguration.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/runtime/CL/gemm/CLGEMMKernelSelection.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLGenerateProposalsLayer.cpp b/src/runtime/CL/functions/CLGenerateProposalsLayer.cpp
index 45dc402..5291de0 100644
--- a/src/runtime/CL/functions/CLGenerateProposalsLayer.cpp
+++ b/src/runtime/CL/functions/CLGenerateProposalsLayer.cpp
@@ -25,6 +25,7 @@
 
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/Types.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLInstanceNormalizationLayer.cpp b/src/runtime/CL/functions/CLInstanceNormalizationLayer.cpp
index fce1fe4..4a60ee9 100644
--- a/src/runtime/CL/functions/CLInstanceNormalizationLayer.cpp
+++ b/src/runtime/CL/functions/CLInstanceNormalizationLayer.cpp
@@ -26,6 +26,8 @@
 #include "arm_compute/core/CL/kernels/CLInstanceNormalizationLayerKernel.h"
 #include "arm_compute/core/Types.h"
 
+#include "support/MemorySupport.h"
+
 namespace arm_compute
 {
 CLInstanceNormalizationLayer::CLInstanceNormalizationLayer()
diff --git a/src/runtime/CL/functions/CLLSTMLayerQuantized.cpp b/src/runtime/CL/functions/CLLSTMLayerQuantized.cpp
index e30b1db..76a531b 100644
--- a/src/runtime/CL/functions/CLLSTMLayerQuantized.cpp
+++ b/src/runtime/CL/functions/CLLSTMLayerQuantized.cpp
@@ -27,6 +27,7 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 #include <memory>
 
diff --git a/src/runtime/CL/functions/CLPriorBoxLayer.cpp b/src/runtime/CL/functions/CLPriorBoxLayer.cpp
index 1907c7c..fefbff6 100644
--- a/src/runtime/CL/functions/CLPriorBoxLayer.cpp
+++ b/src/runtime/CL/functions/CLPriorBoxLayer.cpp
@@ -31,6 +31,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 
+#include "support/MemorySupport.h"
+
 using namespace arm_compute;
 
 CLPriorBoxLayer::CLPriorBoxLayer()
diff --git a/src/runtime/CL/functions/CLQLSTMLayer.cpp b/src/runtime/CL/functions/CLQLSTMLayer.cpp
index 15a54c7..c493471 100644
--- a/src/runtime/CL/functions/CLQLSTMLayer.cpp
+++ b/src/runtime/CL/functions/CLQLSTMLayer.cpp
@@ -30,6 +30,7 @@
 #include "arm_compute/core/utils/misc/InfoHelpers.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLReduceMean.cpp b/src/runtime/CL/functions/CLReduceMean.cpp
index 0e2ede7..4ea7f76 100644
--- a/src/runtime/CL/functions/CLReduceMean.cpp
+++ b/src/runtime/CL/functions/CLReduceMean.cpp
@@ -23,12 +23,13 @@
  */
 #include "arm_compute/runtime/CL/functions/CLReduceMean.h"
 
-#include "arm_compute/core/CL/CLValidate.h"
 #include "arm_compute/core/CL/ICLTensor.h"
 #include "arm_compute/core/CL/kernels/CLReductionOperationKernel.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/CL/CLValidate.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLReductionOperation.cpp b/src/runtime/CL/functions/CLReductionOperation.cpp
index 54e91fb..208371c 100644
--- a/src/runtime/CL/functions/CLReductionOperation.cpp
+++ b/src/runtime/CL/functions/CLReductionOperation.cpp
@@ -30,7 +30,9 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
-#include "arm_compute/runtime/Utils.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/runtime/Utils.h"
+
 #include "support/MemorySupport.h"
 
 namespace arm_compute
@@ -47,7 +49,7 @@
     ARM_COMPUTE_RETURN_ERROR_ON_MSG(axis >= TensorShape::num_max_dimensions, "Reduction axis greater than max number of dimensions");
     ARM_COMPUTE_RETURN_ERROR_ON_MSG(axis > 3, "Unsupported reduction axis");
 
-    const unsigned int num_of_stages       = calculate_number_of_stages_only_x_axis(input->dimension(0), axis);
+    const unsigned int num_of_stages       = utils::calculate_number_of_stages_only_x_axis(input->dimension(0), axis);
     const bool         is_serial           = needs_serialized_reduction(op, input->data_type(), axis);
     const bool         is_reshape_required = !keep_dims;
 
@@ -194,7 +196,7 @@
 void CLReductionOperation::configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, unsigned int axis, ReductionOperation op, bool keep_dims)
 {
     ARM_COMPUTE_ERROR_ON_NULLPTR(input, output);
-    _num_of_stages       = calculate_number_of_stages_only_x_axis(input->info()->dimension(0), axis);
+    _num_of_stages       = utils::calculate_number_of_stages_only_x_axis(input->info()->dimension(0), axis);
     _reduction_axis      = axis;
     _is_serial           = needs_serialized_reduction(op, input->info()->data_type(), axis);
     _is_reshape_required = !keep_dims;
diff --git a/src/runtime/CL/functions/CLRemap.cpp b/src/runtime/CL/functions/CLRemap.cpp
index 60b72c5..1e3d614 100644
--- a/src/runtime/CL/functions/CLRemap.cpp
+++ b/src/runtime/CL/functions/CLRemap.cpp
@@ -42,7 +42,7 @@
 
 void CLRemap::configure(const CLCompileContext &compile_context, ICLTensor *input, const ICLTensor *map_x, const ICLTensor *map_y, ICLTensor *output, InterpolationPolicy policy,
                         BorderMode border_mode,
-                        uint8_t constant_border_value)
+                        uint8_t    constant_border_value)
 {
     ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::U8);
     ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(output, 1, DataType::U8);
diff --git a/src/runtime/CL/functions/CLSelect.cpp b/src/runtime/CL/functions/CLSelect.cpp
index c7d7df7..ef80108 100644
--- a/src/runtime/CL/functions/CLSelect.cpp
+++ b/src/runtime/CL/functions/CLSelect.cpp
@@ -27,6 +27,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
 
+#include "support/MemorySupport.h"
+
 using namespace arm_compute;
 
 namespace arm_compute
diff --git a/src/runtime/CL/functions/CLSoftmaxLayer.cpp b/src/runtime/CL/functions/CLSoftmaxLayer.cpp
index 720f911..759c870 100644
--- a/src/runtime/CL/functions/CLSoftmaxLayer.cpp
+++ b/src/runtime/CL/functions/CLSoftmaxLayer.cpp
@@ -31,6 +31,7 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
+#include "src/core/helpers/SoftmaxHelpers.h"
 
 namespace arm_compute
 {
@@ -63,7 +64,7 @@
     {
         _memory_group.manage(&_input_permuted);
         _memory_group.manage(&_output_permuted);
-        _permute_input.configure(compile_context, input, &_input_permuted, get_permutation_vector_from_softmax_axis(actual_axis));
+        _permute_input.configure(compile_context, input, &_input_permuted, softmax_helpers::get_permutation_vector_from_softmax_axis(actual_axis));
         tmp_output = &_output_permuted;
     }
 
@@ -99,7 +100,7 @@
     _sum.allocator()->allocate();
     if(_needs_permute)
     {
-        _permute_output.configure(compile_context, &_output_permuted, output, get_permutation_vector_from_softmax_axis(actual_axis));
+        _permute_output.configure(compile_context, &_output_permuted, output, softmax_helpers::get_permutation_vector_from_softmax_axis(actual_axis));
         _input_permuted.allocator()->allocate();
         _output_permuted.allocator()->allocate();
     }
@@ -117,7 +118,7 @@
     const bool   needs_permute = actual_axis != 0;
     if(needs_permute)
     {
-        const PermutationVector permutation_vector = get_permutation_vector_from_softmax_axis(actual_axis);
+        const PermutationVector permutation_vector = softmax_helpers::get_permutation_vector_from_softmax_axis(actual_axis);
         const TensorShape       permuted_shape     = misc::shape_calculator::compute_permutation_output_shape(*input, permutation_vector);
         TensorInfo              input_permuted(input->clone()->set_tensor_shape(permuted_shape));
         ARM_COMPUTE_RETURN_ON_ERROR(CLPermute::validate(input, &input_permuted, permutation_vector));
diff --git a/src/runtime/CL/functions/CLSplit.cpp b/src/runtime/CL/functions/CLSplit.cpp
index db0b14b..0b27371 100644
--- a/src/runtime/CL/functions/CLSplit.cpp
+++ b/src/runtime/CL/functions/CLSplit.cpp
@@ -30,6 +30,7 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/CL/CLScheduler.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/functions/CLWinogradConvolutionLayer.cpp b/src/runtime/CL/functions/CLWinogradConvolutionLayer.cpp
index 09a35a6..7ad017f 100644
--- a/src/runtime/CL/functions/CLWinogradConvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLWinogradConvolutionLayer.cpp
@@ -102,7 +102,7 @@
 }
 
 void CLWinogradConvolutionLayer::configure(const CLCompileContext &compile_context, ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output,
-                                           const PadStrideInfo &conv_info,
+                                           const PadStrideInfo       &conv_info,
                                            const ActivationLayerInfo &act_info, bool enable_fast_math)
 {
     // Get indices for the width and height
diff --git a/arm_compute/runtime/CL/gemm/CLGEMMKernelSelection.h b/src/runtime/CL/gemm/CLGEMMKernelSelection.h
similarity index 85%
rename from arm_compute/runtime/CL/gemm/CLGEMMKernelSelection.h
rename to src/runtime/CL/gemm/CLGEMMKernelSelection.h
index a6bc008..f6fad7e 100644
--- a/arm_compute/runtime/CL/gemm/CLGEMMKernelSelection.h
+++ b/src/runtime/CL/gemm/CLGEMMKernelSelection.h
@@ -21,15 +21,15 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_CLGEMMKERNELSELECTION_H
-#define ARM_COMPUTE_CLGEMMKERNELSELECTION_H
+#ifndef SRC_CLGEMMKERNELSELECTION_H
+#define SRC_CLGEMMKERNELSELECTION_H
 
 #include "arm_compute/runtime/CL/ICLGEMMKernelSelection.h"
-#include "arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.h"
-#include "arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.h"
-#include "arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionValhall.h"
+#include "src/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.h"
+#include "src/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.h"
+#include "src/runtime/CL/gemm/CLGEMMKernelSelectionValhall.h"
 
-#include <memory>
+#include "support/MemorySupport.h"
 
 namespace arm_compute
 {
@@ -62,4 +62,4 @@
 };
 } // namespace cl_gemm
 } // namespace arm_compute
-#endif /*ARM_COMPUTE_CLGEMMKERNELSELECTION_H */
+#endif /* SRC_CLGEMMKERNELSELECTION_H */
diff --git a/src/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.cpp b/src/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.cpp
index b1dd690..73b9056 100644
--- a/src/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.cpp
+++ b/src/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.cpp
@@ -21,11 +21,11 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.h"
+#include "src/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
 
 #include <map>
 #include <utility>
diff --git a/arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.h b/src/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.h
similarity index 93%
rename from arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.h
rename to src/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.h
index 579bbe3..a495b48 100644
--- a/arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.h
+++ b/src/runtime/CL/gemm/CLGEMMKernelSelectionBifrost.h
@@ -21,8 +21,8 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_CLGEMMKERNELSELECTIONBIFROST_H
-#define ARM_COMPUTE_CLGEMMKERNELSELECTIONBIFROST_H
+#ifndef SRC_CLGEMMKERNELSELECTIONBIFROST_H
+#define SRC_CLGEMMKERNELSELECTIONBIFROST_H
 
 #include "arm_compute/runtime/CL/ICLGEMMKernelSelection.h"
 
@@ -52,4 +52,4 @@
 };
 } // namespace cl_gemm
 } // namespace arm_compute
-#endif /*ARM_COMPUTE_CLGEMMKERNELSELECTIONBIFROST_H */
+#endif /* SRC_CLGEMMKERNELSELECTIONBIFROST_H */
diff --git a/src/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.cpp b/src/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.cpp
index 324c2f7..d172a82 100644
--- a/src/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.cpp
+++ b/src/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.cpp
@@ -21,12 +21,12 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.h"
+#include "src/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
 #include "arm_compute/core/GPUTarget.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
 
 #include <map>
 #include <utility>
diff --git a/arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.h b/src/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.h
similarity index 92%
rename from arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.h
rename to src/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.h
index 5547731..3f6003f 100644
--- a/arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.h
+++ b/src/runtime/CL/gemm/CLGEMMKernelSelectionMidgard.h
@@ -21,8 +21,8 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_CLGEMMKERNELSELECTIONMIDGARD_H
-#define ARM_COMPUTE_CLGEMMKERNELSELECTIONMIDGARD_H
+#ifndef SRC_CLGEMMKERNELSELECTIONMIDGARD_H
+#define SRC_CLGEMMKERNELSELECTIONMIDGARD_H
 
 #include "arm_compute/runtime/CL/ICLGEMMKernelSelection.h"
 
@@ -50,4 +50,4 @@
 };
 } // namespace cl_gemm
 } // namespace arm_compute
-#endif /*ARM_COMPUTE_CLGEMMKERNELSELECTIONMIDGARD_H */
+#endif /* SRC_CLGEMMKERNELSELECTIONMIDGARD_H */
diff --git a/src/runtime/CL/gemm/CLGEMMKernelSelectionValhall.cpp b/src/runtime/CL/gemm/CLGEMMKernelSelectionValhall.cpp
index c50c7ae..acae0e7 100644
--- a/src/runtime/CL/gemm/CLGEMMKernelSelectionValhall.cpp
+++ b/src/runtime/CL/gemm/CLGEMMKernelSelectionValhall.cpp
@@ -21,11 +21,11 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionValhall.h"
+#include "src/runtime/CL/gemm/CLGEMMKernelSelectionValhall.h"
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernelLibrary.h"
-#include "arm_compute/core/CL/gemm/CLGEMMHelpers.h"
+#include "src/core/CL/gemm/CLGEMMHelpers.h"
 
 #include <map>
 #include <utility>
diff --git a/arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionValhall.h b/src/runtime/CL/gemm/CLGEMMKernelSelectionValhall.h
similarity index 92%
rename from arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionValhall.h
rename to src/runtime/CL/gemm/CLGEMMKernelSelectionValhall.h
index 782ef74..cbea9ea 100644
--- a/arm_compute/runtime/CL/gemm/CLGEMMKernelSelectionValhall.h
+++ b/src/runtime/CL/gemm/CLGEMMKernelSelectionValhall.h
@@ -21,8 +21,8 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_CLGEMMKERNELSELECTIONVALHALL_H
-#define ARM_COMPUTE_CLGEMMKERNELSELECTIONVALHALL_H
+#ifndef SRC_CLGEMMKERNELSELECTIONVALHALL_H
+#define SRC_CLGEMMKERNELSELECTIONVALHALL_H
 
 #include "arm_compute/runtime/CL/ICLGEMMKernelSelection.h"
 
@@ -50,4 +50,4 @@
 };
 } // namespace cl_gemm
 } // namespace arm_compute
-#endif /*ARM_COMPUTE_CLGEMMKERNELSELECTIONVALHALL_H */
+#endif /* SRC_CLGEMMKERNELSELECTIONVALHALL_H */
diff --git a/src/runtime/CL/tuners/BifrostTuner.cpp b/src/runtime/CL/tuners/BifrostTuner.cpp
index 52644bf..a6474c9 100644
--- a/src/runtime/CL/tuners/BifrostTuner.cpp
+++ b/src/runtime/CL/tuners/BifrostTuner.cpp
@@ -25,7 +25,7 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernels.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "support/Cast.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CL/tuners/MidgardTuner.cpp b/src/runtime/CL/tuners/MidgardTuner.cpp
index e49e155..58b0d57 100644
--- a/src/runtime/CL/tuners/MidgardTuner.cpp
+++ b/src/runtime/CL/tuners/MidgardTuner.cpp
@@ -25,7 +25,7 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/CLKernels.h"
-#include "arm_compute/core/utils/misc/Cast.h"
+#include "support/Cast.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/CPP/CPPScheduler.cpp b/src/runtime/CPP/CPPScheduler.cpp
index f017006..e6b0ec2 100644
--- a/src/runtime/CPP/CPPScheduler.cpp
+++ b/src/runtime/CPP/CPPScheduler.cpp
@@ -27,7 +27,8 @@
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/runtime/CPUUtils.h"
+#include "src/runtime/CPUUtils.h"
+#include "support/MemorySupport.h"
 #include "support/Mutex.h"
 
 #include <atomic>
diff --git a/src/runtime/CPP/functions/CPPDetectionOutputLayer.cpp b/src/runtime/CPP/functions/CPPDetectionOutputLayer.cpp
index 9d62733..fdb4c9f 100644
--- a/src/runtime/CPP/functions/CPPDetectionOutputLayer.cpp
+++ b/src/runtime/CPP/functions/CPPDetectionOutputLayer.cpp
@@ -26,6 +26,7 @@
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 #include <list>
 
diff --git a/src/runtime/CPP/functions/CPPDetectionPostProcessLayer.cpp b/src/runtime/CPP/functions/CPPDetectionPostProcessLayer.cpp
index 3507a3a..31f1faf 100644
--- a/src/runtime/CPP/functions/CPPDetectionPostProcessLayer.cpp
+++ b/src/runtime/CPP/functions/CPPDetectionPostProcessLayer.cpp
@@ -26,6 +26,7 @@
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 #include <cstddef>
 #include <ios>
diff --git a/src/runtime/CPUUtils.cpp b/src/runtime/CPUUtils.cpp
index 4d6caae..a7dd464 100644
--- a/src/runtime/CPUUtils.cpp
+++ b/src/runtime/CPUUtils.cpp
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/runtime/CPUUtils.h"
+#include "src/runtime/CPUUtils.h"
 
 #include "arm_compute/core/CPP/CPPTypes.h"
 #include "arm_compute/core/Error.h"
@@ -352,6 +352,10 @@
 
 namespace arm_compute
 {
+namespace utils
+{
+namespace cpu
+{
 void get_cpu_configuration(CPUInfo &cpuinfo)
 {
 #if !defined(BARE_METAL) && (defined(__arm__) || defined(__aarch64__))
@@ -460,5 +464,6 @@
 
     return num_threads_hint;
 }
-
+} // namespace cpu
+} // namespace utils
 } // namespace arm_compute
diff --git a/arm_compute/runtime/CPUUtils.h b/src/runtime/CPUUtils.h
similarity index 92%
rename from arm_compute/runtime/CPUUtils.h
rename to src/runtime/CPUUtils.h
index bcc2f66..452d3d5 100644
--- a/arm_compute/runtime/CPUUtils.h
+++ b/src/runtime/CPUUtils.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -27,6 +27,11 @@
 namespace arm_compute
 {
 class CPUInfo;
+
+namespace utils
+{
+namespace cpu
+{
 /** This function will try to detect the CPU configuration on the system and will fill
  *  the cpuinfo object accordingly to reflect this.
  *
@@ -40,5 +45,7 @@
  * @return The minumum number of common cores.
  */
 unsigned int get_threads_hint();
-}
+} // namespace cpu
+} // namespace utils
+} // namespace arm_compute
 #endif /* ARM_COMPUTE_RUNTIME_CPU_UTILS_H */
diff --git a/src/runtime/DeviceProperties.cpp b/src/runtime/DeviceProperties.cpp
index 5d7ae02..ec9f4a1 100644
--- a/src/runtime/DeviceProperties.cpp
+++ b/src/runtime/DeviceProperties.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,12 +23,12 @@
  */
 #include "arm_compute/runtime/DeviceProperties.h"
 
-#include "arm_compute/runtime/CPUUtils.h"
+#include "src/runtime/CPUUtils.h"
 
 namespace arm_compute
 {
 DeviceProperties::DeviceProperties()
 {
-    get_cpu_configuration(cpu_info);
+    utils::cpu::get_cpu_configuration(cpu_info);
 }
 } // namespace arm_compute
diff --git a/src/runtime/GLES_COMPUTE/GCMemory.cpp b/src/runtime/GLES_COMPUTE/GCMemory.cpp
index 998f8a5..4d74555 100644
--- a/src/runtime/GLES_COMPUTE/GCMemory.cpp
+++ b/src/runtime/GLES_COMPUTE/GCMemory.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -23,8 +23,8 @@
  */
 #include "arm_compute/runtime/GLES_COMPUTE/GCMemory.h"
 
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/runtime/GLES_COMPUTE/GCMemoryRegion.h"
+#include "support/Cast.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/GLES_COMPUTE/functions/GCConcatenateLayer.cpp b/src/runtime/GLES_COMPUTE/functions/GCConcatenateLayer.cpp
index 9e23974..807412e 100644
--- a/src/runtime/GLES_COMPUTE/functions/GCConcatenateLayer.cpp
+++ b/src/runtime/GLES_COMPUTE/functions/GCConcatenateLayer.cpp
@@ -29,6 +29,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/GLES_COMPUTE/GCScheduler.h"
+#include "src/core/helpers/AutoConfiguration.h"
+
 #include "support/MemorySupport.h"
 
 namespace arm_compute
diff --git a/src/runtime/IScheduler.cpp b/src/runtime/IScheduler.cpp
index 53df369..43df3d5 100644
--- a/src/runtime/IScheduler.cpp
+++ b/src/runtime/IScheduler.cpp
@@ -26,17 +26,17 @@
 #include "arm_compute/core/CPP/ICPPKernel.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Window.h"
-#include "arm_compute/runtime/CPUUtils.h"
-#include "arm_compute/runtime/SchedulerUtils.h"
+#include "src/runtime/CPUUtils.h"
+#include "src/runtime/SchedulerUtils.h"
 
 namespace arm_compute
 {
 IScheduler::IScheduler()
     : _cpu_info()
 {
-    get_cpu_configuration(_cpu_info);
+    utils::cpu::get_cpu_configuration(_cpu_info);
     // Work out the best possible number of execution threads
-    _num_threads_hint = get_threads_hint();
+    _num_threads_hint = utils::cpu::get_threads_hint();
 }
 
 CPUInfo &IScheduler::cpu_info()
@@ -74,7 +74,7 @@
 
         //in c++17 this can be swapped for   auto [ m_threads, n_threads ] = split_2d(...
         unsigned m_threads, n_threads;
-        std::tie(m_threads, n_threads) = split_2d(this->num_threads(), m, n);
+        std::tie(m_threads, n_threads) = scheduler_utils::split_2d(this->num_threads(), m, n);
 
         std::vector<IScheduler::Workload> workloads;
         for(unsigned int ni = 0; ni != n_threads; ++ni)
diff --git a/src/runtime/NEON/INESimpleFunctionNoBorder.cpp b/src/runtime/NEON/INESimpleFunctionNoBorder.cpp
index 82316c4..f2181e0 100644
--- a/src/runtime/NEON/INESimpleFunctionNoBorder.cpp
+++ b/src/runtime/NEON/INESimpleFunctionNoBorder.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -24,7 +24,7 @@
 #include "arm_compute/runtime/NEON/INESimpleFunctionNoBorder.h"
 
 #include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "arm_compute/runtime/Utils.h"
+#include "src/runtime/Utils.h"
 
 namespace arm_compute
 {
@@ -36,6 +36,6 @@
 
 void INESimpleFunctionNoBorder::run()
 {
-    schedule_kernel_on_ctx(_ctx, _kernel.get(), Window::DimY);
+    utils::schedule_kernel_on_ctx(_ctx, _kernel.get(), Window::DimY);
 }
 } // namespace arm_compute
diff --git a/src/runtime/NEON/functions/NEArgMinMaxLayer.cpp b/src/runtime/NEON/functions/NEArgMinMaxLayer.cpp
index 0664d3c..70bbba6 100644
--- a/src/runtime/NEON/functions/NEArgMinMaxLayer.cpp
+++ b/src/runtime/NEON/functions/NEArgMinMaxLayer.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 
+#include "support/MemorySupport.h"
+
 namespace arm_compute
 {
 NEArgMinMaxLayer::NEArgMinMaxLayer(std::shared_ptr<IMemoryManager> memory_manager)
diff --git a/src/runtime/NEON/functions/NEBatchNormalizationLayer.cpp b/src/runtime/NEON/functions/NEBatchNormalizationLayer.cpp
index 5a593e9..eab40ac 100644
--- a/src/runtime/NEON/functions/NEBatchNormalizationLayer.cpp
+++ b/src/runtime/NEON/functions/NEBatchNormalizationLayer.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2018 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -30,6 +30,8 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 
+#include "support/MemorySupport.h"
+
 using namespace arm_compute;
 
 NEBatchNormalizationLayer::NEBatchNormalizationLayer()
diff --git a/src/runtime/NEON/functions/NEBatchToSpaceLayer.cpp b/src/runtime/NEON/functions/NEBatchToSpaceLayer.cpp
index c06a8aa..2705cff 100644
--- a/src/runtime/NEON/functions/NEBatchToSpaceLayer.cpp
+++ b/src/runtime/NEON/functions/NEBatchToSpaceLayer.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,6 +29,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 
+#include "support/MemorySupport.h"
+
 namespace arm_compute
 {
 void NEBatchToSpaceLayer::configure(const ITensor *input, const ITensor *block_shape, ITensor *output)
diff --git a/src/runtime/NEON/functions/NEConcatenateLayer.cpp b/src/runtime/NEON/functions/NEConcatenateLayer.cpp
index 8df4f4c..72bd9e6 100644
--- a/src/runtime/NEON/functions/NEConcatenateLayer.cpp
+++ b/src/runtime/NEON/functions/NEConcatenateLayer.cpp
@@ -35,6 +35,7 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Types.h"
+#include "src/core/helpers/AutoConfiguration.h"
 #include "support/MemorySupport.h"
 
 namespace arm_compute
diff --git a/src/runtime/NEON/functions/NECropResize.cpp b/src/runtime/NEON/functions/NECropResize.cpp
index f6ed2ec..f8f9916 100644
--- a/src/runtime/NEON/functions/NECropResize.cpp
+++ b/src/runtime/NEON/functions/NECropResize.cpp
@@ -25,6 +25,8 @@
 
 #include "arm_compute/runtime/NEON/functions/NECropResize.h"
 
+#include "support/MemorySupport.h"
+
 #include <cstddef>
 
 namespace arm_compute
diff --git a/src/runtime/NEON/functions/NEDeconvolutionLayer.cpp b/src/runtime/NEON/functions/NEDeconvolutionLayer.cpp
index dff3070..cb9ab16 100644
--- a/src/runtime/NEON/functions/NEDeconvolutionLayer.cpp
+++ b/src/runtime/NEON/functions/NEDeconvolutionLayer.cpp
@@ -28,6 +28,7 @@
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 using namespace arm_compute::misc::shape_calculator;
 
diff --git a/src/runtime/NEON/functions/NEDepthToSpaceLayer.cpp b/src/runtime/NEON/functions/NEDepthToSpaceLayer.cpp
index e363f89..0aaa37e 100644
--- a/src/runtime/NEON/functions/NEDepthToSpaceLayer.cpp
+++ b/src/runtime/NEON/functions/NEDepthToSpaceLayer.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -29,6 +29,8 @@
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Validate.h"
 
+#include "support/MemorySupport.h"
+
 namespace arm_compute
 {
 void NEDepthToSpaceLayer::configure(const ITensor *input, ITensor *output, int32_t block_shape)
diff --git a/src/runtime/NEON/functions/NEFFT1D.cpp b/src/runtime/NEON/functions/NEFFT1D.cpp
index 744a915..2c53b18 100644
--- a/src/runtime/NEON/functions/NEFFT1D.cpp
+++ b/src/runtime/NEON/functions/NEFFT1D.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -25,8 +25,8 @@
 
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/utils/helpers/fft.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
+#include "src/core/utils/helpers/fft.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEFFTConvolutionLayer.cpp b/src/runtime/NEON/functions/NEFFTConvolutionLayer.cpp
index cd68788..a46fc9f 100644
--- a/src/runtime/NEON/functions/NEFFTConvolutionLayer.cpp
+++ b/src/runtime/NEON/functions/NEFFTConvolutionLayer.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -26,8 +26,11 @@
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
-#include "arm_compute/core/utils/helpers/fft.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/utils/helpers/fft.h"
+
+#include "support/MemorySupport.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEFullyConnectedLayer.cpp b/src/runtime/NEON/functions/NEFullyConnectedLayer.cpp
index 4dcf41e..d956d16 100644
--- a/src/runtime/NEON/functions/NEFullyConnectedLayer.cpp
+++ b/src/runtime/NEON/functions/NEFullyConnectedLayer.cpp
@@ -30,6 +30,8 @@
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 
+#include "support/MemorySupport.h"
+
 #include <algorithm>
 #include <cmath>
 
diff --git a/src/runtime/NEON/functions/NEGEMM.cpp b/src/runtime/NEON/functions/NEGEMM.cpp
index 3b8ca44..4166cff 100644
--- a/src/runtime/NEON/functions/NEGEMM.cpp
+++ b/src/runtime/NEON/functions/NEGEMM.cpp
@@ -23,7 +23,6 @@
  */
 #include "arm_compute/runtime/NEON/functions/NEGEMM.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
@@ -34,6 +33,8 @@
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 #include "arm_compute/runtime/NEON/functions/NEGEMMAssemblyDispatch.h"
 #include "arm_compute/runtime/TensorAllocator.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 #include <cmath>
 
diff --git a/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp b/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp
index ad349cb..5b08483 100644
--- a/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp
+++ b/src/runtime/NEON/functions/NEGEMMAssemblyDispatch.cpp
@@ -23,13 +23,13 @@
  */
 #include "arm_compute/runtime/NEON/functions/NEGEMMAssemblyDispatch.h"
 
+#include "arm_compute/runtime/NEON/NEScheduler.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h"
+#include "src/core/NEON/kernels/assembly/NEGEMMAssemblyWrapperKernel.h"
 #include "src/core/NEON/kernels/assembly/arm_gemm.hpp"
 
-#include "arm_compute/core/CPP/Validate.h"
-#include "arm_compute/runtime/NEON/NEScheduler.h"
-#include "arm_compute/runtime/NEON/functions/NESimpleAssemblyFunction.h"
-
-#include "src/core/NEON/kernels/assembly/NEGEMMAssemblyWrapperKernel.h"
+#include "support/MemorySupport.h"
 
 #include <arm_neon.h>
 
diff --git a/src/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.cpp b/src/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.cpp
index 83db146..36357dd 100644
--- a/src/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.cpp
+++ b/src/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.cpp
@@ -33,6 +33,7 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 #include "arm_compute/runtime/TensorAllocator.h"
+#include "src/core/helpers/AutoConfiguration.h"
 #include "support/MemorySupport.h"
 
 namespace arm_compute
diff --git a/src/runtime/NEON/functions/NEGenerateProposalsLayer.cpp b/src/runtime/NEON/functions/NEGenerateProposalsLayer.cpp
index 3d53778..13210a0 100644
--- a/src/runtime/NEON/functions/NEGenerateProposalsLayer.cpp
+++ b/src/runtime/NEON/functions/NEGenerateProposalsLayer.cpp
@@ -25,6 +25,7 @@
 
 #include "arm_compute/core/Types.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NELSTMLayerQuantized.cpp b/src/runtime/NEON/functions/NELSTMLayerQuantized.cpp
index 11989d3..7610d15 100644
--- a/src/runtime/NEON/functions/NELSTMLayerQuantized.cpp
+++ b/src/runtime/NEON/functions/NELSTMLayerQuantized.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -26,6 +26,7 @@
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 #include <cmath>
 #include <memory>
diff --git a/src/runtime/NEON/functions/NEPadLayer.cpp b/src/runtime/NEON/functions/NEPadLayer.cpp
index 21c349b..03c597a 100644
--- a/src/runtime/NEON/functions/NEPadLayer.cpp
+++ b/src/runtime/NEON/functions/NEPadLayer.cpp
@@ -27,6 +27,7 @@
 
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEPriorBoxLayer.cpp b/src/runtime/NEON/functions/NEPriorBoxLayer.cpp
index fda130b..bcf6bef 100644
--- a/src/runtime/NEON/functions/NEPriorBoxLayer.cpp
+++ b/src/runtime/NEON/functions/NEPriorBoxLayer.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -31,6 +31,8 @@
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 
+#include "support/MemorySupport.h"
+
 namespace arm_compute
 {
 void NEPriorBoxLayer::configure(const ITensor *input1, const ITensor *input2, ITensor *output, const PriorBoxLayerInfo &info)
diff --git a/src/runtime/NEON/functions/NEQLSTMLayer.cpp b/src/runtime/NEON/functions/NEQLSTMLayer.cpp
index 5a6b513..95f20ae 100644
--- a/src/runtime/NEON/functions/NEQLSTMLayer.cpp
+++ b/src/runtime/NEON/functions/NEQLSTMLayer.cpp
@@ -30,6 +30,7 @@
 #include "arm_compute/core/utils/misc/InfoHelpers.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
+#include "src/core/helpers/WindowHelpers.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEReduceMean.cpp b/src/runtime/NEON/functions/NEReduceMean.cpp
index 021f7b5..c3c5529 100644
--- a/src/runtime/NEON/functions/NEReduceMean.cpp
+++ b/src/runtime/NEON/functions/NEReduceMean.cpp
@@ -23,11 +23,12 @@
  */
 #include "arm_compute/runtime/NEON/functions/NEReduceMean.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEReductionOperation.cpp b/src/runtime/NEON/functions/NEReductionOperation.cpp
index 91176bf..4938a56 100644
--- a/src/runtime/NEON/functions/NEReductionOperation.cpp
+++ b/src/runtime/NEON/functions/NEReductionOperation.cpp
@@ -26,6 +26,7 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
+#include "src/core/helpers/AutoConfiguration.h"
 
 namespace arm_compute
 {
diff --git a/src/runtime/NEON/functions/NEScale.cpp b/src/runtime/NEON/functions/NEScale.cpp
index 2278f07..bbf8343 100644
--- a/src/runtime/NEON/functions/NEScale.cpp
+++ b/src/runtime/NEON/functions/NEScale.cpp
@@ -30,12 +30,14 @@
 #include "arm_compute/core/PixelValue.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/misc/Rounding.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 #include "arm_compute/runtime/TensorAllocator.h"
 
 #include "src/core/utils/ScaleUtils.h"
 
+#include "support/MemorySupport.h"
+#include "support/Rounding.h"
+
 #include <cmath>
 #include <cstddef>
 #include <utility>
diff --git a/src/runtime/NEON/functions/NESimpleAssemblyFunction.cpp b/src/runtime/NEON/functions/NESimpleAssemblyFunction.cpp
index b0cafae..d165b22 100644
--- a/src/runtime/NEON/functions/NESimpleAssemblyFunction.cpp
+++ b/src/runtime/NEON/functions/NESimpleAssemblyFunction.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/runtime/NEON/functions/NESimpleAssemblyFunction.h"
+#include "src/runtime/NEON/functions/NESimpleAssemblyFunction.h"
 
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
diff --git a/arm_compute/runtime/NEON/functions/NESimpleAssemblyFunction.h b/src/runtime/NEON/functions/NESimpleAssemblyFunction.h
similarity index 94%
rename from arm_compute/runtime/NEON/functions/NESimpleAssemblyFunction.h
rename to src/runtime/NEON/functions/NESimpleAssemblyFunction.h
index a814802..e9be54d 100644
--- a/arm_compute/runtime/NEON/functions/NESimpleAssemblyFunction.h
+++ b/src/runtime/NEON/functions/NESimpleAssemblyFunction.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_NESIMPLEASSEMBLYFUNCTION_H
 #define ARM_COMPUTE_NESIMPLEASSEMBLYFUNCTION_H
 
-#include "arm_compute/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h"
 #include "arm_compute/runtime/IFunction.h"
+#include "src/core/NEON/kernels/assembly/INEGEMMWrapperKernel.h"
 
 #include <memory>
 
diff --git a/src/runtime/NEON/functions/NESoftmaxLayer.cpp b/src/runtime/NEON/functions/NESoftmaxLayer.cpp
index e763caa..4f77386 100644
--- a/src/runtime/NEON/functions/NESoftmaxLayer.cpp
+++ b/src/runtime/NEON/functions/NESoftmaxLayer.cpp
@@ -27,6 +27,7 @@
 #include "arm_compute/core/NEON/kernels/NESoftmaxLayerKernel.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
+#include "src/core/helpers/SoftmaxHelpers.h"
 
 namespace arm_compute
 {
@@ -53,7 +54,7 @@
         // Add to the memory manager _input_permuted
         _memory_group.manage(&_input_permuted);
 
-        _permute_input.configure(input, &_input_permuted, get_permutation_vector_from_softmax_axis(actual_axis));
+        _permute_input.configure(input, &_input_permuted, softmax_helpers::get_permutation_vector_from_softmax_axis(actual_axis));
     }
 
     // We want to deal with a 2D input. Either it is the permuted version of the original input (4D case)
@@ -87,7 +88,7 @@
         _input_permuted.allocator()->allocate();
 
         // Re-permute the permuted output into the requested (4D) output
-        _permute_output.configure(&_output_permuted, output, get_permutation_vector_from_softmax_axis(actual_axis));
+        _permute_output.configure(&_output_permuted, output, softmax_helpers::get_permutation_vector_from_softmax_axis(actual_axis));
 
         // Allocate the intermediate permuted tensors
         _output_permuted.allocator()->allocate();
@@ -128,7 +129,7 @@
 
     if(needs_permute)
     {
-        const PermutationVector permutation_vector = get_permutation_vector_from_softmax_axis(actual_axis);
+        const PermutationVector permutation_vector = softmax_helpers::get_permutation_vector_from_softmax_axis(actual_axis);
         const TensorShape       permuted_shape     = misc::shape_calculator::compute_permutation_output_shape(*input, permutation_vector);
         TensorInfo              input_permuted(input->clone()->set_tensor_shape(permuted_shape));
         ARM_COMPUTE_RETURN_ON_ERROR(NEPermute::validate(input, &input_permuted, permutation_vector));
diff --git a/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp b/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp
index 1bad310..23b9f60 100644
--- a/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp
+++ b/src/runtime/NEON/functions/NEWinogradConvolutionLayer.cpp
@@ -23,17 +23,17 @@
  */
 #include "arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 #include "arm_compute/runtime/NEON/functions/NEGEMMAssemblyDispatch.h"
+#include "src/core/CPP/Validate.h"
 #include "src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.h"
 #include "support/MemorySupport.h"
 
-#include "arm_compute/core/NEON/kernels/convolution/common/utils.hpp"
+#include "src/core/NEON/kernels/convolution/common/utils.hpp"
 #include "src/core/NEON/kernels/convolution/winograd/winograd.hpp"
 
 namespace arm_compute
diff --git a/src/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.cpp b/src/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.cpp
index 73a7caa..11e89cb 100644
--- a/src/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.cpp
+++ b/src/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.cpp
@@ -24,18 +24,21 @@
 
 #include "arm_compute/runtime/NEON/functions/assembly/NEDepthwiseConvolutionAssemblyDispatch.h"
 
-#include "arm_compute/core/CPP/Validate.h"
 #include "arm_compute/core/ITensor.h"
-#include "arm_compute/core/NEON/kernels/assembly/NEDepthwiseConvolutionAssemblyKernelWrapper.h"
-#include "arm_compute/core/NEON/kernels/convolution/depthwise/depthwise_dilated.hpp"
-#include "arm_compute/core/NEON/kernels/convolution/depthwise/depthwise_quantized_dilated.hpp"
 #include "arm_compute/core/Utils.h"
 #include "arm_compute/core/utils/misc/InfoHelpers.h"
 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "src/core/CPP/Validate.h"
+#include "src/core/NEON/kernels/assembly/NEDepthwiseConvolutionAssemblyKernelWrapper.h"
+#include "src/core/NEON/kernels/convolution/depthwise/depthwise_dilated.hpp"
+#include "src/core/NEON/kernels/convolution/depthwise/depthwise_quantized_dilated.hpp"
+#include "src/core/helpers/AutoConfiguration.h"
 
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 
+#include "support/MemorySupport.h"
+
 #include <set>
 
 namespace arm_compute
diff --git a/src/runtime/OMP/OMPScheduler.cpp b/src/runtime/OMP/OMPScheduler.cpp
index 4c2f03a..bf34b01 100644
--- a/src/runtime/OMP/OMPScheduler.cpp
+++ b/src/runtime/OMP/OMPScheduler.cpp
@@ -27,7 +27,7 @@
 #include "arm_compute/core/Error.h"
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Utils.h"
-#include "arm_compute/runtime/CPUUtils.h"
+#include "src/runtime/CPUUtils.h"
 #include <omp.h>
 
 namespace arm_compute
diff --git a/src/runtime/SchedulerUtils.cpp b/src/runtime/SchedulerUtils.cpp
index 1c12e3c..6f9a32c 100644
--- a/src/runtime/SchedulerUtils.cpp
+++ b/src/runtime/SchedulerUtils.cpp
@@ -21,6 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
+#include "src/runtime/SchedulerUtils.h"
 
 #include "arm_compute/core/Error.h"
 
@@ -28,6 +29,8 @@
 
 namespace arm_compute
 {
+namespace scheduler_utils
+{
 #ifndef BARE_METAL
 std::pair<unsigned, unsigned> split_2d(unsigned max_threads, std::size_t m, std::size_t n)
 {
@@ -76,4 +79,5 @@
     }
 }
 #endif /* #ifndef BARE_METAL */
+} // namespace scheduler_utils
 } // namespace arm_compute
diff --git a/arm_compute/runtime/SchedulerUtils.h b/src/runtime/SchedulerUtils.h
similarity index 80%
rename from arm_compute/runtime/SchedulerUtils.h
rename to src/runtime/SchedulerUtils.h
index 5a88e03..46644a3 100644
--- a/arm_compute/runtime/SchedulerUtils.h
+++ b/src/runtime/SchedulerUtils.h
@@ -21,13 +21,18 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_SCHEDULER_UTILS_H
-#define ARM_COMPUTE_SCHEDULER_UTILS_H
+#ifndef SRC_COMPUTE_SCHEDULER_UTILS_H
+#define SRC_COMPUTE_SCHEDULER_UTILS_H
+
+#include <cstddef>
+#include <utility>
 
 namespace arm_compute
 {
-/** Given two dimensions and a maxium number of threads to utilise, calculate the best
- * combination of threads that fit in (mutliplied together) max_threads.
+namespace scheduler_utils
+{
+/** Given two dimensions and a maximum number of threads to utilise, calculate the best
+ * combination of threads that fit in (multiplied together) max_threads.
  *
  * This algorithm assumes that work in either of the dimensions is equally difficult
  * to compute
@@ -35,5 +40,6 @@
  * @returns [m_nthreads, n_nthreads] A pair of the threads that should be used in each dimension
  */
 std::pair<unsigned, unsigned> split_2d(unsigned max_threads, std::size_t m, std::size_t n);
+} // namespace scheduler_utils
 } // namespace arm_compute
-#endif /* ARM_COMPUTE_SCHEDULER_UTILS_H */
+#endif /* SRC_COMPUTE_SCHEDULER_UTILS_H */
diff --git a/src/runtime/Utils.cpp b/src/runtime/Utils.cpp
index 534b421..15e9d43 100644
--- a/src/runtime/Utils.cpp
+++ b/src/runtime/Utils.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,7 +21,7 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#include "arm_compute/runtime/Utils.h"
+#include "src/runtime/Utils.h"
 
 #include "arm_compute/runtime/NEON/NEScheduler.h"
 
@@ -31,6 +31,8 @@
 
 namespace arm_compute
 {
+namespace utils
+{
 #ifndef DOXYGEN_SKIP_THIS
 static const std::string information =
 #include "arm_compute_version.embed"
@@ -78,4 +80,5 @@
     const unsigned int num_of_stages = num_of_wg / 128 + 2;
     return num_of_stages;
 }
+} // namespace utils
 } // namespace arm_compute
diff --git a/arm_compute/runtime/Utils.h b/src/runtime/Utils.h
similarity index 92%
rename from arm_compute/runtime/Utils.h
rename to src/runtime/Utils.h
index 6e36297..f8775c9 100644
--- a/arm_compute/runtime/Utils.h
+++ b/src/runtime/Utils.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,8 +21,8 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_RUNTIME_UTILS_H
-#define ARM_COMPUTE_RUNTIME_UTILS_H
+#ifndef SRC_RUNTIME_UTILS_H
+#define SRC_RUNTIME_UTILS_H
 
 #include "arm_compute/runtime/IRuntimeContext.h"
 #include "arm_compute/runtime/Scheduler.h"
@@ -31,6 +31,8 @@
 
 namespace arm_compute
 {
+namespace utils
+{
 /** Convert a Scheduler::Type into a string.
  *
  * @param[in] t @ref Scheduler::Type to be translated to string.
@@ -53,5 +55,6 @@
  * @param[in] axis              axis to be used
  */
 unsigned int calculate_number_of_stages_only_x_axis(size_t input_x_dimension, unsigned int axis);
+} // namespace utils
 } // namespace arm_compute
-#endif /* ARM_COMPUTE_RUNTIME_UTILS_H */
+#endif /* SRC_RUNTIME_UTILS_H */
diff --git a/arm_compute/core/utils/misc/CRTP.h b/support/CRTP.h
similarity index 97%
rename from arm_compute/core/utils/misc/CRTP.h
rename to support/CRTP.h
index d295500..eb358d0 100644
--- a/arm_compute/core/utils/misc/CRTP.h
+++ b/support/CRTP.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/arm_compute/core/utils/misc/Cast.h b/support/Cast.h
similarity index 98%
rename from arm_compute/core/utils/misc/Cast.h
rename to support/Cast.h
index 57c7e49..53d5f68 100644
--- a/arm_compute/core/utils/misc/Cast.h
+++ b/support/Cast.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/arm_compute/core/utils/misc/ICloneable.h b/support/ICloneable.h
similarity index 97%
rename from arm_compute/core/utils/misc/ICloneable.h
rename to support/ICloneable.h
index cbb0b3c..5d333c8 100644
--- a/arm_compute/core/utils/misc/ICloneable.h
+++ b/support/ICloneable.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/arm_compute/core/utils/misc/Iterable.h b/support/Iterable.h
similarity index 98%
rename from arm_compute/core/utils/misc/Iterable.h
rename to support/Iterable.h
index 3423208..a0bafaf 100644
--- a/arm_compute/core/utils/misc/Iterable.h
+++ b/support/Iterable.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/arm_compute/core/utils/misc/Random.h b/support/Random.h
similarity index 98%
rename from arm_compute/core/utils/misc/Random.h
rename to support/Random.h
index 6832c49..c8b767e 100644
--- a/arm_compute/core/utils/misc/Random.h
+++ b/support/Random.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 Arm Limited.
+ * Copyright (c) 2019-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/arm_compute/core/utils/misc/Requires.h b/support/Requires.h
similarity index 97%
rename from arm_compute/core/utils/misc/Requires.h
rename to support/Requires.h
index ba91039..bc4932a 100644
--- a/arm_compute/core/utils/misc/Requires.h
+++ b/support/Requires.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/arm_compute/core/utils/misc/Rounding.h b/support/Rounding.h
similarity index 98%
rename from arm_compute/core/utils/misc/Rounding.h
rename to support/Rounding.h
index 1ed4e64..ba9266d 100644
--- a/arm_compute/core/utils/misc/Rounding.h
+++ b/support/Rounding.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -25,8 +25,8 @@
 #define ARM_COMPUTE_UTILS_ROUNDING_H
 
 #include "arm_compute/core/Error.h"
-#include "arm_compute/core/utils/misc/Requires.h"
 #include "arm_compute/core/utils/misc/Traits.h"
+#include "support/Requires.h"
 #include "support/ToolchainSupport.h"
 
 #include <cmath>
diff --git a/arm_compute/core/utils/misc/SaturateCast.h b/support/SaturateCast.h
similarity index 98%
rename from arm_compute/core/utils/misc/SaturateCast.h
rename to support/SaturateCast.h
index cbced83..a9982d8 100644
--- a/arm_compute/core/utils/misc/SaturateCast.h
+++ b/support/SaturateCast.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -24,9 +24,9 @@
 #ifndef ARM_COMPUTE_UTILS_CAST_SATURATE_CAST_H
 #define ARM_COMPUTE_UTILS_CAST_SATURATE_CAST_H
 
-#include "arm_compute/core/utils/misc/Rounding.h"
 #include "arm_compute/core/utils/misc/Traits.h"
 #include "arm_compute/core/utils/misc/Utility.h"
+#include "support/Rounding.h"
 
 namespace arm_compute
 {
diff --git a/arm_compute/core/utils/misc/CRTP.h b/support/Traits.h
similarity index 64%
copy from arm_compute/core/utils/misc/CRTP.h
copy to support/Traits.h
index d295500..e892c63 100644
--- a/arm_compute/core/utils/misc/CRTP.h
+++ b/support/Traits.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+* Copyright (c) 2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,35 +21,27 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_MISC_CRTP_H
-#define ARM_COMPUTE_MISC_CRTP_H
+#ifndef SUPPORT_TRAITS_H
+#define SUPPORT_TRAITS_H
+
+#include <type_traits>
 
 namespace arm_compute
 {
-namespace misc
+/** Disable bitwise operations by default */
+template <typename T>
+struct enable_bitwise_ops
 {
-/** Curiously recurring template pattern Interface */
-template <typename T, template <typename> class Type>
-struct CRTP
-{
-public:
-    /** Exact type */
-    using ExactType = T;
-
-protected:
-    const T &impl() const
-    {
-        return static_cast<const T &>(*this);
-    }
-    T &impl()
-    {
-        return static_cast<T &>(*this);
-    }
-
-private:
-    CRTP() = default;
-    friend Type<T>;
+    static constexpr bool value = false; /**< Disabled */
 };
-} // namespace misc
+
+#ifndef DOXYGEN_SKIP_THIS
+template <typename T>
+typename std::enable_if<enable_bitwise_ops<T>::value, T>::type operator&(T lhs, T rhs)
+{
+    using underlying_type = typename std::underlying_type<T>::type;
+    return static_cast<T>(static_cast<underlying_type>(lhs) & static_cast<underlying_type>(rhs));
+}
+#endif /* DOXYGEN_SKIP_THIS */
 } // namespace arm_compute
-#endif /* ARM_COMPUTE_MISC_CRTP_H */
+#endif /* SUPPORT_TRAITS_H */
diff --git a/tests/AssetsLibrary.h b/tests/AssetsLibrary.h
index d783b1f..28a51d6 100644
--- a/tests/AssetsLibrary.h
+++ b/tests/AssetsLibrary.h
@@ -31,7 +31,7 @@
 #include "arm_compute/core/TensorShape.h"
 #include "arm_compute/core/Types.h"
 #include "arm_compute/core/Window.h"
-#include "arm_compute/core/utils/misc/Random.h"
+#include "support/Random.h"
 #include "tests/RawTensor.h"
 #include "tests/TensorCache.h"
 #include "tests/Utils.h"
diff --git a/tests/framework/instruments/SchedulerTimer.cpp b/tests/framework/instruments/SchedulerTimer.cpp
index aa69bc2..b4d1c59 100644
--- a/tests/framework/instruments/SchedulerTimer.cpp
+++ b/tests/framework/instruments/SchedulerTimer.cpp
@@ -26,8 +26,8 @@
 #include "Instruments.h"
 #include "WallClockTimer.h"
 #include "arm_compute/core/CPP/ICPPKernel.h"
-#include "arm_compute/core/utils/misc/Cast.h"
 #include "arm_compute/graph/INode.h"
+#include "support/Cast.h"
 
 namespace arm_compute
 {
diff --git a/tests/validation/NEON/ActivationLayer.cpp b/tests/validation/NEON/ActivationLayer.cpp
index 9b2cad1..b294ff6 100644
--- a/tests/validation/NEON/ActivationLayer.cpp
+++ b/tests/validation/NEON/ActivationLayer.cpp
@@ -22,7 +22,6 @@
  * SOFTWARE.
  */
 #include "arm_compute/core/Types.h"
-#include "arm_compute/core/utils/misc/Requires.h"
 #include "arm_compute/core/utils/misc/Traits.h"
 #include "arm_compute/runtime/NEON/functions/NEActivationLayer.h"
 #include "arm_compute/runtime/RuntimeContext.h"
@@ -38,6 +37,8 @@
 #include "tests/validation/Validation.h"
 #include "tests/validation/fixtures/ActivationLayerFixture.h"
 
+#include "support/Requires.h"
+
 namespace arm_compute
 {
 namespace test
diff --git a/tests/validation/reference/Convolution3d.h b/tests/validation/reference/Convolution3d.h
index 03a2f53..34e27f4 100644
--- a/tests/validation/reference/Convolution3d.h
+++ b/tests/validation/reference/Convolution3d.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2019 Arm Limited.
+ * Copyright (c) 2017-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -24,8 +24,8 @@
 #ifndef ARM_COMPUTE_TEST_VALIDATION_CONVOLUTION_H
 #define ARM_COMPUTE_TEST_VALIDATION_CONVOLUTION_H
 
-#include "arm_compute/core/utils/misc/Requires.h"
 #include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "support/Requires.h"
 #include "tests/validation/Helpers.h"
 #include "tests/validation/reference/UtilsQuantizedAsymm.h"
 
diff --git a/tests/validation/reference/DepthConvertLayer.cpp b/tests/validation/reference/DepthConvertLayer.cpp
index 30b7e57..94c719a 100644
--- a/tests/validation/reference/DepthConvertLayer.cpp
+++ b/tests/validation/reference/DepthConvertLayer.cpp
@@ -25,8 +25,8 @@
 
 #include "tests/validation/Helpers.h"
 
-#include "arm_compute/core/utils/misc/Rounding.h"
-#include "arm_compute/core/utils/misc/SaturateCast.h"
+#include "support/Rounding.h"
+#include "support/SaturateCast.h"
 
 #include "tests/Types.h"
 
diff --git a/tests/validation/reference/Scale.cpp b/tests/validation/reference/Scale.cpp
index aa265c2..71e98fd 100644
--- a/tests/validation/reference/Scale.cpp
+++ b/tests/validation/reference/Scale.cpp
@@ -25,9 +25,9 @@
 #include "Scale.h"
 
 #include "Utils.h"
-#include "arm_compute/core/utils/misc/Rounding.h"
 #include "arm_compute/core/utils/misc/Utility.h"
 #include "src/core/utils/ScaleUtils.h"
+#include "support/Rounding.h"
 
 namespace arm_compute
 {
diff --git a/tests/validation/reference/SliceOperations.cpp b/tests/validation/reference/SliceOperations.cpp
index 50c5c68..222b146 100644
--- a/tests/validation/reference/SliceOperations.cpp
+++ b/tests/validation/reference/SliceOperations.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2019 Arm Limited.
+ * Copyright (c) 2018-2020 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
diff --git a/tests/validation/reference/UpsampleLayer.cpp b/tests/validation/reference/UpsampleLayer.cpp
index 4e06ad4..e5eb376 100644
--- a/tests/validation/reference/UpsampleLayer.cpp
+++ b/tests/validation/reference/UpsampleLayer.cpp
@@ -23,7 +23,7 @@
  */
 #include "UpsampleLayer.h"
 
-#include "arm_compute/core/utils/misc/Requires.h"
+#include "support/Requires.h"
 #include "tests/validation/Helpers.h"
 
 namespace arm_compute
diff --git a/utils/Utils.h b/utils/Utils.h
index 1941d37..e44d978 100644
--- a/utils/Utils.h
+++ b/utils/Utils.h
@@ -31,7 +31,6 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensor.h"
 #include "arm_compute/core/Types.h"
-#include "arm_compute/core/Validate.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/runtime/Tensor.h"
 #pragma GCC diagnostic push