diff --git a/src/core/NEON/kernels/NELogicalKernel.cpp b/src/core/NEON/kernels/NELogicalKernel.cpp
index 27605e1..d98694f 100644
--- a/src/core/NEON/kernels/NELogicalKernel.cpp
+++ b/src/core/NEON/kernels/NELogicalKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020 Arm Limited.
+ * Copyright (c) 2020-2021 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -25,7 +25,7 @@
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
-#include "src/core/common/Validate.h"
+#include "src/common/utils/Validate.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
 
diff --git a/src/core/NEON/kernels/batchnormalization/impl/NEON/fp16.cpp b/src/core/NEON/kernels/batchnormalization/impl/NEON/fp16.cpp
index 4108d79..c105ada 100644
--- a/src/core/NEON/kernels/batchnormalization/impl/NEON/fp16.cpp
+++ b/src/core/NEON/kernels/batchnormalization/impl/NEON/fp16.cpp
@@ -27,7 +27,6 @@
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/kernels/detail/NEActivationFunctionDetail.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 
 #include <arm_neon.h>
 #include <cmath>
diff --git a/src/core/NEON/kernels/batchnormalization/impl/NEON/fp32.cpp b/src/core/NEON/kernels/batchnormalization/impl/NEON/fp32.cpp
index 1fdc5bd..4a90a21 100644
--- a/src/core/NEON/kernels/batchnormalization/impl/NEON/fp32.cpp
+++ b/src/core/NEON/kernels/batchnormalization/impl/NEON/fp32.cpp
@@ -27,7 +27,6 @@
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/kernels/detail/NEActivationFunctionDetail.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 
 #include <arm_neon.h>
 #include <cmath>
diff --git a/src/core/NEON/kernels/batchnormalization/impl/SVE/fp16.cpp b/src/core/NEON/kernels/batchnormalization/impl/SVE/fp16.cpp
index 5638dce..3e3e81d 100644
--- a/src/core/NEON/kernels/batchnormalization/impl/SVE/fp16.cpp
+++ b/src/core/NEON/kernels/batchnormalization/impl/SVE/fp16.cpp
@@ -25,7 +25,6 @@
 #include "arm_compute/core/ITensorPack.h"
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/SVEMath.h"
-#include "src/core/common/Validate.h"
 
 #include <cmath>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/batchnormalization/impl/SVE/fp32.cpp b/src/core/NEON/kernels/batchnormalization/impl/SVE/fp32.cpp
index 51397ac..b0d4cbb 100644
--- a/src/core/NEON/kernels/batchnormalization/impl/SVE/fp32.cpp
+++ b/src/core/NEON/kernels/batchnormalization/impl/SVE/fp32.cpp
@@ -25,7 +25,6 @@
 #include "arm_compute/core/ITensorPack.h"
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/SVEMath.h"
-#include "src/core/common/Validate.h"
 
 #include <cmath>
 #include <cstddef>
diff --git a/src/core/NEON/kernels/scale/impl/NEON/fp16.cpp b/src/core/NEON/kernels/scale/impl/NEON/fp16.cpp
new file mode 100644
index 0000000..0ad66ca
--- /dev/null
+++ b/src/core/NEON/kernels/scale/impl/NEON/fp16.cpp
@@ -0,0 +1,174 @@
+/*
+ * Copyright (c) 2021 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 "arm_compute/core/Helpers.h"
+#include "arm_compute/core/ITensorPack.h"
+#include "arm_compute/core/Window.h"
+#include "src/core/NEON/NEMath.h"
+#include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/ScaleHelpers.h"
+#include "src/core/utils/ScaleUtils.h"
+#include "support/Rounding.h"
+
+#include <arm_neon.h>
+#include <cmath>
+#include <cstddef>
+
+#if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(ENABLE_FP16_KERNELS)
+
+namespace arm_compute
+{
+namespace
+{
+void fp16_neon_scale_nearest(const ITensor *src, ITensor *dst, const ITensor *offsets,
+                             float sampling_offset, bool align_corners, const Window &window)
+{
+    const size_t in_stride_c  = src->info()->dimension(0) + src->info()->padding().left + src->info()->padding().right;
+    const size_t in_stride_w  = src->info()->dimension(1) + src->info()->padding().top + src->info()->padding().bottom;
+    const size_t in_stride_wc = in_stride_w * in_stride_c;
+    const size_t in_dim_h     = src->info()->dimension(2);
+
+    // Compute the ratio between source height and destination height
+    const auto hr             = scale_utils::calculate_resize_ratio(in_dim_h, dst->info()->dimension(2), align_corners);
+    const auto window_start_x = static_cast<int32_t>(window.x().start());
+    const auto window_end_x   = static_cast<int32_t>(window.x().end());
+    const int  window_step_x  = 8;
+
+    Window win(window);
+    win.set(Window::DimX, Window::Dimension(0, 1, 1));
+    Iterator out(dst, win);
+
+    const uint8_t     *in_ptr_start        = src->buffer() + src->info()->offset_first_element_in_bytes();
+    const unsigned int in_stride_bytes_hwc = src->info()->strides_in_bytes()[3];
+
+    execute_window_loop(win, [&](const Coordinates & id)
+    {
+        const int32_t    offset     = *reinterpret_cast<const int32_t *>(offsets->ptr_to_element(Coordinates(id.y(), id.z()))) * in_stride_c;
+        const auto       in_hi      = static_cast<int>(align_corners ? utils::rounding::round_half_away_from_zero((id.z() + sampling_offset) * hr) : std::floor((id.z() + sampling_offset) * hr));
+        const int        offset_row = in_hi * in_stride_wc;
+        int32_t          x          = window_start_x;
+        const float16_t *in_ptr     = reinterpret_cast<const float16_t *>(in_ptr_start + in_stride_bytes_hwc * id[3]);
+
+        for(; x <= window_end_x - window_step_x; x += window_step_x)
+        {
+            wrapper::vstore(reinterpret_cast<float16_t *>(out.ptr()) + x,
+                            wrapper::vloadq(in_ptr + offset + offset_row + x));
+        }
+        for(; x < window_end_x; ++x)
+        {
+            *(reinterpret_cast<float16_t *>(out.ptr()) + x) = *(in_ptr + offset + offset_row + x);
+        }
+    },
+    out);
+}
+
+void fp16_neon_scale_bilinear(const ITensor *src, ITensor *dst, const ITensor *offsets, const ITensor *dx, const ITensor *dy,
+                              BorderMode border_mode, PixelValue constant_border_value, float sampling_offset,
+                              bool align_corners, const Window &window)
+{
+    // Compute the ratio between source height and destination height
+    const auto hr = scale_utils::calculate_resize_ratio(src->info()->dimension(2), dst->info()->dimension(2), align_corners);
+
+    Iterator  out(dst, window);
+    const int in_stride_c  = src->info()->dimension(0) + src->info()->padding().left + src->info()->padding().right;
+    const int in_dim_w     = src->info()->dimension(1);
+    const int in_dim_h     = src->info()->dimension(2);
+    const int in_stride_wc = in_stride_c * (in_dim_w + src->info()->padding().top + src->info()->padding().bottom);
+
+    // Don't increment in Y and Z 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);
+    win_in.set(Window::DimY, Window::Dimension(0, 0, 0));
+    win_in.set(Window::DimZ, Window::Dimension(0, 0, 0));
+    Iterator in(src, win_in);
+
+    if(border_mode == BorderMode::CONSTANT)
+    {
+        using ConstType = typename std::conditional<std::is_same<float16_t, float16_t>::value, half, float16_t>::type;
+
+        const float16_t const_border_value = static_cast<float16_t>(constant_border_value.get<ConstType>());
+        execute_window_loop(window, [&](const Coordinates & id)
+        {
+            const auto       offset = *reinterpret_cast<const int32_t *>(offsets->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto       dx_val = *reinterpret_cast<const float *>(dx->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto       dy_val = *reinterpret_cast<const float *>(dy->ptr_to_element(Coordinates(id.y(), id.z())));
+            const int32_t    in_hi  = std::floor((id.z() + sampling_offset) * hr - sampling_offset);
+            const float16_t *in_ptr = reinterpret_cast<const float16_t *>(in.ptr()) + offset * in_stride_c + in_hi * in_stride_wc;
+
+            const auto a00 = (0 <= offset && offset < in_dim_w && 0 <= in_hi && in_hi < in_dim_h) ? *in_ptr : const_border_value;
+            const auto a01 = (-1 <= offset && offset < in_dim_w - 1 && 0 <= in_hi && in_hi < in_dim_h) ? *(in_ptr + in_stride_c) : const_border_value;
+            const auto a10 = (0 <= offset && offset < in_dim_w && -1 <= in_hi && in_hi < in_dim_h - 1) ? *(in_ptr + in_stride_wc) : const_border_value;
+            const auto a11 = (-1 <= offset && offset < in_dim_w - 1 && -1 <= in_hi && in_hi < in_dim_h - 1) ? *(in_ptr + in_stride_c + in_stride_wc) : const_border_value;
+
+            *reinterpret_cast<float16_t *>(out.ptr()) = static_cast<float16_t>(scale_helpers::delta_bilinear(a00, a01, a10, a11, dx_val, dy_val));
+        },
+        in, out);
+    }
+    else if(border_mode == BorderMode::REPLICATE)
+    {
+        execute_window_loop(window, [&](const Coordinates & id)
+        {
+            const auto offset = *reinterpret_cast<const int32_t *>(offsets->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto dx_val = *reinterpret_cast<const float *>(dx->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto dy_val = *reinterpret_cast<const float *>(dy->ptr_to_element(Coordinates(id.y(), id.z())));
+            const int  in_hi  = std::floor((id.z() + sampling_offset) * hr - sampling_offset);
+
+            auto clamped_w  = utility::clamp<int>(offset, 0, in_dim_w - 1);
+            auto clamped_w1 = utility::clamp<int>(offset + 1, 0, in_dim_w - 1);
+            auto clamped_h  = utility::clamp<int>(in_hi, 0, in_dim_h - 1);
+            auto clamped_h1 = utility::clamp<int>(in_hi + 1, 0, in_dim_h - 1);
+
+            const auto a00 = *(reinterpret_cast<const float16_t *>(in.ptr()) + clamped_w * in_stride_c + clamped_h * in_stride_wc);
+            const auto a01 = *(reinterpret_cast<const float16_t *>(in.ptr()) + clamped_w1 * in_stride_c + clamped_h * in_stride_wc);
+            const auto a10 = *(reinterpret_cast<const float16_t *>(in.ptr()) + clamped_w * in_stride_c + clamped_h1 * in_stride_wc);
+            const auto a11 = *(reinterpret_cast<const float16_t *>(in.ptr()) + clamped_w1 * in_stride_c + clamped_h1 * in_stride_wc);
+
+            *reinterpret_cast<float16_t *>(out.ptr()) = static_cast<float16_t>(scale_helpers::delta_bilinear(a00, a01, a10, a11, dx_val, dy_val));
+        },
+        in, out);
+    }
+    else
+    {
+        ARM_COMPUTE_ERROR("Not implemented");
+    }
+}
+}
+namespace cpu
+{
+void fp16_neon_scale(const ITensor *src, ITensor *dst, const ITensor *offsets, const ITensor *dx, const ITensor *dy,
+                     InterpolationPolicy policy, BorderMode border_mode, PixelValue constant_border_value, float sampling_offset,
+                     bool align_corners, const Window &window)
+{
+    if(policy == InterpolationPolicy::BILINEAR)
+    {
+        fp16_neon_scale_bilinear(src, dst, offsets, dx, dy, border_mode, constant_border_value, sampling_offset, align_corners, window);
+    }
+    else if(policy == InterpolationPolicy::NEAREST_NEIGHBOR)
+    {
+        fp16_neon_scale_nearest(src, dst, offsets, sampling_offset, align_corners, window);
+    }
+}
+} // namespace cpu
+} // namespace arm_compute
+
+#endif /* defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(ENABLE_FP16_KERNELS) */
\ No newline at end of file
diff --git a/src/core/NEON/kernels/scale/impl/NEON/integer.cpp b/src/core/NEON/kernels/scale/impl/NEON/integer.cpp
new file mode 100644
index 0000000..a2359aa
--- /dev/null
+++ b/src/core/NEON/kernels/scale/impl/NEON/integer.cpp
@@ -0,0 +1,293 @@
+/*
+ * Copyright (c) 2021 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 "arm_compute/core/Helpers.h"
+#include "arm_compute/core/ITensorPack.h"
+#include "arm_compute/core/Window.h"
+#include "src/core/NEON/NEMath.h"
+#include "src/core/NEON/wrapper/wrapper.h"
+#include "src/core/helpers/ScaleHelpers.h"
+#include "src/core/utils/ScaleUtils.h"
+#include "support/Rounding.h"
+
+#include <arm_neon.h>
+#include <cmath>
+#include <cstddef>
+
+namespace arm_compute
+{
+namespace
+{
+void u8_neon_scale_nearest(const ITensor *src, ITensor *dst, const ITensor *offsets,
+                           float sampling_offset, bool align_corners, const Window &window)
+{
+    const size_t in_stride_c  = src->info()->dimension(0) + src->info()->padding().left + src->info()->padding().right;
+    const size_t in_stride_w  = src->info()->dimension(1) + src->info()->padding().top + src->info()->padding().bottom;
+    const size_t in_stride_wc = in_stride_w * in_stride_c;
+    const size_t in_dim_h     = src->info()->dimension(2);
+
+    // Compute the ratio between source height and destination height
+    const auto hr             = scale_utils::calculate_resize_ratio(in_dim_h, dst->info()->dimension(2), align_corners);
+    const auto window_start_x = static_cast<int32_t>(window.x().start());
+    const auto window_end_x   = static_cast<int32_t>(window.x().end());
+    const int  window_step_x  = 16;
+
+    Window win(window);
+    win.set(Window::DimX, Window::Dimension(0, 1, 1));
+    Iterator out(dst, win);
+
+    const uint8_t     *in_ptr_start        = src->buffer() + src->info()->offset_first_element_in_bytes();
+    const unsigned int in_stride_bytes_hwc = src->info()->strides_in_bytes()[3];
+
+    execute_window_loop(win, [&](const Coordinates & id)
+    {
+        const int32_t  offset     = *reinterpret_cast<const int32_t *>(offsets->ptr_to_element(Coordinates(id.y(), id.z()))) * in_stride_c;
+        const auto     in_hi      = static_cast<int>(align_corners ? utils::rounding::round_half_away_from_zero((id.z() + sampling_offset) * hr) : std::floor((id.z() + sampling_offset) * hr));
+        const int      offset_row = in_hi * in_stride_wc;
+        int32_t        x          = window_start_x;
+        const uint8_t *in_ptr     = reinterpret_cast<const uint8_t *>(in_ptr_start + in_stride_bytes_hwc * id[3]);
+
+        for(; x <= window_end_x - window_step_x; x += window_step_x)
+        {
+            wrapper::vstore(reinterpret_cast<uint8_t *>(out.ptr()) + x,
+                            wrapper::vloadq(in_ptr + offset + offset_row + x));
+        }
+        for(; x < window_end_x; ++x)
+        {
+            *(reinterpret_cast<uint8_t *>(out.ptr()) + x) = *(in_ptr + offset + offset_row + x);
+        }
+    },
+    out);
+}
+
+void u8_neon_scale_bilinear(const ITensor *src, ITensor *dst, const ITensor *offsets, const ITensor *dx, const ITensor *dy,
+                            BorderMode border_mode, PixelValue constant_border_value, float sampling_offset,
+                            bool align_corners, const Window &window)
+{
+    // Compute the ratio between source height and destination height
+    const auto hr = scale_utils::calculate_resize_ratio(src->info()->dimension(2), dst->info()->dimension(2), align_corners);
+
+    Iterator  out(dst, window);
+    const int in_stride_c  = src->info()->dimension(0) + src->info()->padding().left + src->info()->padding().right;
+    const int in_dim_w     = src->info()->dimension(1);
+    const int in_dim_h     = src->info()->dimension(2);
+    const int in_stride_wc = in_stride_c * (in_dim_w + src->info()->padding().top + src->info()->padding().bottom);
+
+    // Don't increment in Y and Z 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);
+    win_in.set(Window::DimY, Window::Dimension(0, 0, 0));
+    win_in.set(Window::DimZ, Window::Dimension(0, 0, 0));
+    Iterator in(src, win_in);
+
+    if(border_mode == BorderMode::CONSTANT)
+    {
+        const uint8_t const_border_value = static_cast<uint8_t>(constant_border_value.get<uint8_t>());
+        execute_window_loop(window, [&](const Coordinates & id)
+        {
+            const auto     offset = *reinterpret_cast<const int32_t *>(offsets->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto     dx_val = *reinterpret_cast<const float *>(dx->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto     dy_val = *reinterpret_cast<const float *>(dy->ptr_to_element(Coordinates(id.y(), id.z())));
+            const int32_t  in_hi  = std::floor((id.z() + sampling_offset) * hr - sampling_offset);
+            const uint8_t *in_ptr = reinterpret_cast<const uint8_t *>(in.ptr()) + offset * in_stride_c + in_hi * in_stride_wc;
+
+            const auto a00 = (0 <= offset && offset < in_dim_w && 0 <= in_hi && in_hi < in_dim_h) ? *in_ptr : const_border_value;
+            const auto a01 = (-1 <= offset && offset < in_dim_w - 1 && 0 <= in_hi && in_hi < in_dim_h) ? *(in_ptr + in_stride_c) : const_border_value;
+            const auto a10 = (0 <= offset && offset < in_dim_w && -1 <= in_hi && in_hi < in_dim_h - 1) ? *(in_ptr + in_stride_wc) : const_border_value;
+            const auto a11 = (-1 <= offset && offset < in_dim_w - 1 && -1 <= in_hi && in_hi < in_dim_h - 1) ? *(in_ptr + in_stride_c + in_stride_wc) : const_border_value;
+
+            *reinterpret_cast<uint8_t *>(out.ptr()) = static_cast<uint8_t>(scale_helpers::delta_bilinear(a00, a01, a10, a11, dx_val, dy_val));
+        },
+        in, out);
+    }
+    else if(border_mode == BorderMode::REPLICATE)
+    {
+        execute_window_loop(window, [&](const Coordinates & id)
+        {
+            const auto offset = *reinterpret_cast<const int32_t *>(offsets->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto dx_val = *reinterpret_cast<const float *>(dx->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto dy_val = *reinterpret_cast<const float *>(dy->ptr_to_element(Coordinates(id.y(), id.z())));
+            const int  in_hi  = std::floor((id.z() + sampling_offset) * hr - sampling_offset);
+
+            auto clamped_w  = utility::clamp<int>(offset, 0, in_dim_w - 1);
+            auto clamped_w1 = utility::clamp<int>(offset + 1, 0, in_dim_w - 1);
+            auto clamped_h  = utility::clamp<int>(in_hi, 0, in_dim_h - 1);
+            auto clamped_h1 = utility::clamp<int>(in_hi + 1, 0, in_dim_h - 1);
+
+            const auto a00 = *(reinterpret_cast<const uint8_t *>(in.ptr()) + clamped_w * in_stride_c + clamped_h * in_stride_wc);
+            const auto a01 = *(reinterpret_cast<const uint8_t *>(in.ptr()) + clamped_w1 * in_stride_c + clamped_h * in_stride_wc);
+            const auto a10 = *(reinterpret_cast<const uint8_t *>(in.ptr()) + clamped_w * in_stride_c + clamped_h1 * in_stride_wc);
+            const auto a11 = *(reinterpret_cast<const uint8_t *>(in.ptr()) + clamped_w1 * in_stride_c + clamped_h1 * in_stride_wc);
+
+            *reinterpret_cast<uint8_t *>(out.ptr()) = static_cast<uint8_t>(scale_helpers::delta_bilinear(a00, a01, a10, a11, dx_val, dy_val));
+        },
+        in, out);
+    }
+    else
+    {
+        ARM_COMPUTE_ERROR("Not implemented");
+    }
+}
+
+void s16_neon_scale_nearest(const ITensor *src, ITensor *dst, const ITensor *offsets,
+                            float sampling_offset, bool align_corners, const Window &window)
+{
+    const size_t in_stride_c  = src->info()->dimension(0) + src->info()->padding().left + src->info()->padding().right;
+    const size_t in_stride_w  = src->info()->dimension(1) + src->info()->padding().top + src->info()->padding().bottom;
+    const size_t in_stride_wc = in_stride_w * in_stride_c;
+    const size_t in_dim_h     = src->info()->dimension(2);
+
+    // Compute the ratio between source height and destination height
+    const auto hr             = scale_utils::calculate_resize_ratio(in_dim_h, dst->info()->dimension(2), align_corners);
+    const auto window_start_x = static_cast<int32_t>(window.x().start());
+    const auto window_end_x   = static_cast<int32_t>(window.x().end());
+    const int  window_step_x  = 8;
+
+    Window win(window);
+    win.set(Window::DimX, Window::Dimension(0, 1, 1));
+    Iterator out(dst, win);
+
+    const uint8_t     *in_ptr_start        = src->buffer() + src->info()->offset_first_element_in_bytes();
+    const unsigned int in_stride_bytes_hwc = src->info()->strides_in_bytes()[3];
+
+    execute_window_loop(win, [&](const Coordinates & id)
+    {
+        const int32_t  offset     = *reinterpret_cast<const int32_t *>(offsets->ptr_to_element(Coordinates(id.y(), id.z()))) * in_stride_c;
+        const auto     in_hi      = static_cast<int>(align_corners ? utils::rounding::round_half_away_from_zero((id.z() + sampling_offset) * hr) : std::floor((id.z() + sampling_offset) * hr));
+        const int      offset_row = in_hi * in_stride_wc;
+        int32_t        x          = window_start_x;
+        const int16_t *in_ptr     = reinterpret_cast<const int16_t *>(in_ptr_start + in_stride_bytes_hwc * id[3]);
+
+        for(; x <= window_end_x - window_step_x; x += window_step_x)
+        {
+            wrapper::vstore(reinterpret_cast<int16_t *>(out.ptr()) + x,
+                            wrapper::vloadq(in_ptr + offset + offset_row + x));
+        }
+        for(; x < window_end_x; ++x)
+        {
+            *(reinterpret_cast<int16_t *>(out.ptr()) + x) = *(in_ptr + offset + offset_row + x);
+        }
+    },
+    out);
+}
+
+void s16_neon_scale_bilinear(const ITensor *src, ITensor *dst, const ITensor *offsets, const ITensor *dx, const ITensor *dy,
+                             BorderMode border_mode, PixelValue constant_border_value, float sampling_offset,
+                             bool align_corners, const Window &window)
+{
+    // Compute the ratio between source height and destination height
+    const auto hr = scale_utils::calculate_resize_ratio(src->info()->dimension(2), dst->info()->dimension(2), align_corners);
+
+    Iterator  out(dst, window);
+    const int in_stride_c  = src->info()->dimension(0) + src->info()->padding().left + src->info()->padding().right;
+    const int in_dim_w     = src->info()->dimension(1);
+    const int in_dim_h     = src->info()->dimension(2);
+    const int in_stride_wc = in_stride_c * (in_dim_w + src->info()->padding().top + src->info()->padding().bottom);
+
+    // Don't increment in Y and Z 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);
+    win_in.set(Window::DimY, Window::Dimension(0, 0, 0));
+    win_in.set(Window::DimZ, Window::Dimension(0, 0, 0));
+    Iterator in(src, win_in);
+
+    if(border_mode == BorderMode::CONSTANT)
+    {
+        const int16_t const_border_value = static_cast<int16_t>(constant_border_value.get<int16_t>());
+        execute_window_loop(window, [&](const Coordinates & id)
+        {
+            const auto     offset = *reinterpret_cast<const int32_t *>(offsets->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto     dx_val = *reinterpret_cast<const float *>(dx->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto     dy_val = *reinterpret_cast<const float *>(dy->ptr_to_element(Coordinates(id.y(), id.z())));
+            const int32_t  in_hi  = std::floor((id.z() + sampling_offset) * hr - sampling_offset);
+            const int16_t *in_ptr = reinterpret_cast<const int16_t *>(in.ptr()) + offset * in_stride_c + in_hi * in_stride_wc;
+
+            const auto a00 = (0 <= offset && offset < in_dim_w && 0 <= in_hi && in_hi < in_dim_h) ? *in_ptr : const_border_value;
+            const auto a01 = (-1 <= offset && offset < in_dim_w - 1 && 0 <= in_hi && in_hi < in_dim_h) ? *(in_ptr + in_stride_c) : const_border_value;
+            const auto a10 = (0 <= offset && offset < in_dim_w && -1 <= in_hi && in_hi < in_dim_h - 1) ? *(in_ptr + in_stride_wc) : const_border_value;
+            const auto a11 = (-1 <= offset && offset < in_dim_w - 1 && -1 <= in_hi && in_hi < in_dim_h - 1) ? *(in_ptr + in_stride_c + in_stride_wc) : const_border_value;
+
+            *reinterpret_cast<int16_t *>(out.ptr()) = static_cast<int16_t>(scale_helpers::delta_bilinear(a00, a01, a10, a11, dx_val, dy_val));
+        },
+        in, out);
+    }
+    else if(border_mode == BorderMode::REPLICATE)
+    {
+        execute_window_loop(window, [&](const Coordinates & id)
+        {
+            const auto offset = *reinterpret_cast<const int32_t *>(offsets->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto dx_val = *reinterpret_cast<const float *>(dx->ptr_to_element(Coordinates(id.y(), id.z())));
+            const auto dy_val = *reinterpret_cast<const float *>(dy->ptr_to_element(Coordinates(id.y(), id.z())));
+            const int  in_hi  = std::floor((id.z() + sampling_offset) * hr - sampling_offset);
+
+            auto clamped_w  = utility::clamp<int>(offset, 0, in_dim_w - 1);
+            auto clamped_w1 = utility::clamp<int>(offset + 1, 0, in_dim_w - 1);
+            auto clamped_h  = utility::clamp<int>(in_hi, 0, in_dim_h - 1);
+            auto clamped_h1 = utility::clamp<int>(in_hi + 1, 0, in_dim_h - 1);
+
+            const auto a00 = *(reinterpret_cast<const int16_t *>(in.ptr()) + clamped_w * in_stride_c + clamped_h * in_stride_wc);
+            const auto a01 = *(reinterpret_cast<const int16_t *>(in.ptr()) + clamped_w1 * in_stride_c + clamped_h * in_stride_wc);
+            const auto a10 = *(reinterpret_cast<const int16_t *>(in.ptr()) + clamped_w * in_stride_c + clamped_h1 * in_stride_wc);
+            const auto a11 = *(reinterpret_cast<const int16_t *>(in.ptr()) + clamped_w1 * in_stride_c + clamped_h1 * in_stride_wc);
+
+            *reinterpret_cast<int16_t *>(out.ptr()) = static_cast<int16_t>(scale_helpers::delta_bilinear(a00, a01, a10, a11, dx_val, dy_val));
+        },
+        in, out);
+    }
+    else
+    {
+        ARM_COMPUTE_ERROR("Not implemented");
+    }
+}
+}
+namespace cpu
+{
+void u8_neon_scale(const ITensor *src, ITensor *dst, const ITensor *offsets, const ITensor *dx, const ITensor *dy,
+                   InterpolationPolicy policy, BorderMode border_mode, PixelValue constant_border_value, float sampling_offset,
+                   bool align_corners, const Window &window)
+{
+    if(policy == InterpolationPolicy::BILINEAR)
+    {
+        u8_neon_scale_bilinear(src, dst, offsets, dx, dy, border_mode, constant_border_value, sampling_offset, align_corners, window);
+    }
+    else if(policy == InterpolationPolicy::NEAREST_NEIGHBOR)
+    {
+        u8_neon_scale_nearest(src, dst, offsets, sampling_offset, align_corners, window);
+    }
+}
+
+void s16_neon_scale(const ITensor *src, ITensor *dst, const ITensor *offsets, const ITensor *dx, const ITensor *dy,
+                    InterpolationPolicy policy, BorderMode border_mode, PixelValue constant_border_value, float sampling_offset,
+                    bool align_corners, const Window &window)
+{
+    if(policy == InterpolationPolicy::BILINEAR)
+    {
+        s16_neon_scale_bilinear(src, dst, offsets, dx, dy, border_mode, constant_border_value, sampling_offset, align_corners, window);
+    }
+    else if(policy == InterpolationPolicy::NEAREST_NEIGHBOR)
+    {
+        s16_neon_scale_nearest(src, dst, offsets, sampling_offset, align_corners, window);
+    }
+}
+} // namespace cpu
+} // namespace arm_compute
\ No newline at end of file
diff --git a/src/core/NEON/kernels/scale/impl/NEON/list.h b/src/core/NEON/kernels/scale/impl/NEON/list.h
index 11cac26..c91242f 100644
--- a/src/core/NEON/kernels/scale/impl/NEON/list.h
+++ b/src/core/NEON/kernels/scale/impl/NEON/list.h
@@ -29,7 +29,6 @@
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 #include "src/core/helpers/ScaleHelpers.h"
 #include "src/core/utils/ScaleUtils.h"
 #include "support/Rounding.h"
diff --git a/src/core/NEON/kernels/scale/impl/NEON/qasymm8_signed.cpp b/src/core/NEON/kernels/scale/impl/NEON/qasymm8_signed.cpp
index 7b18608..149cdf4 100644
--- a/src/core/NEON/kernels/scale/impl/NEON/qasymm8_signed.cpp
+++ b/src/core/NEON/kernels/scale/impl/NEON/qasymm8_signed.cpp
@@ -21,7 +21,6 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-
 #include "src/core/NEON/kernels/scale/impl/NEON/list.h"
 
 namespace arm_compute
diff --git a/src/core/NEON/kernels/scale/impl/SVE/fp16.cpp b/src/core/NEON/kernels/scale/impl/SVE/fp16.cpp
index 91c3dc3..99f08db 100644
--- a/src/core/NEON/kernels/scale/impl/SVE/fp16.cpp
+++ b/src/core/NEON/kernels/scale/impl/SVE/fp16.cpp
@@ -26,7 +26,6 @@
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 #include "src/core/helpers/ScaleHelpers.h"
 #include "src/core/utils/ScaleUtils.h"
 #include "support/Rounding.h"
diff --git a/src/core/NEON/kernels/scale/impl/SVE/fp32.cpp b/src/core/NEON/kernels/scale/impl/SVE/fp32.cpp
index abb4faa..94055ae 100644
--- a/src/core/NEON/kernels/scale/impl/SVE/fp32.cpp
+++ b/src/core/NEON/kernels/scale/impl/SVE/fp32.cpp
@@ -26,7 +26,6 @@
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 #include "src/core/helpers/ScaleHelpers.h"
 #include "src/core/utils/ScaleUtils.h"
 #include "support/Rounding.h"
diff --git a/src/core/NEON/kernels/scale/impl/SVE/integer.cpp b/src/core/NEON/kernels/scale/impl/SVE/integer.cpp
index 5f5263c..2a724ec 100644
--- a/src/core/NEON/kernels/scale/impl/SVE/integer.cpp
+++ b/src/core/NEON/kernels/scale/impl/SVE/integer.cpp
@@ -26,7 +26,6 @@
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 #include "src/core/helpers/ScaleHelpers.h"
 #include "src/core/utils/ScaleUtils.h"
 #include "support/Rounding.h"
diff --git a/src/core/NEON/kernels/scale/impl/SVE/qasymm8.cpp b/src/core/NEON/kernels/scale/impl/SVE/qasymm8.cpp
index fc65ff4..c475ad6 100644
--- a/src/core/NEON/kernels/scale/impl/SVE/qasymm8.cpp
+++ b/src/core/NEON/kernels/scale/impl/SVE/qasymm8.cpp
@@ -26,7 +26,6 @@
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 #include "src/core/helpers/ScaleHelpers.h"
 #include "src/core/helpers/ScaleHelpers.h"
 #include "src/core/utils/ScaleUtils.h"
diff --git a/src/core/NEON/kernels/scale/impl/SVE/qasymm8_signed.cpp b/src/core/NEON/kernels/scale/impl/SVE/qasymm8_signed.cpp
index 676ca94..b39b75a 100644
--- a/src/core/NEON/kernels/scale/impl/SVE/qasymm8_signed.cpp
+++ b/src/core/NEON/kernels/scale/impl/SVE/qasymm8_signed.cpp
@@ -26,7 +26,6 @@
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 #include "src/core/helpers/ScaleHelpers.h"
 #include "src/core/helpers/ScaleHelpers.h"
 #include "src/core/utils/ScaleUtils.h"
diff --git a/src/core/common/Validate.h b/src/core/common/Validate.h
deleted file mode 100644
index fa24bf5..0000000
--- a/src/core/common/Validate.h
+++ /dev/null
@@ -1,41 +0,0 @@
-/*
- * 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_COMMON_VALIDATE_H
-#define SRC_CORE_COMMON_VALIDATE_H
-
-#if defined(ARM_COMPUTE_ASSERTS_ENABLED)
-
-#include <cassert>
-
-#define ARM_COMPUTE_ASSERT(cond) assert(cond)
-#define ARM_COMPUTE_ASSERT_NOT_NULLPTR(ptr) assert((ptr) != nullptr)
-
-#else /* defined(ARM_COMPUTE_ASSERTS_ENABLED) */
-
-#define ARM_COMPUTE_ASSERT(cond)
-#define ARM_COMPUTE_ASSERT_NOT_NULLPTR(ptr)
-
-#endif /* defined(ARM_COMPUTE_ASSERTS_ENABLED) */
-#endif /* SRC_CORE_COMMON_VALIDATE_H */
diff --git a/src/core/cpu/kernels/activation/NEON/fp16.cpp b/src/core/cpu/kernels/activation/NEON/fp16.cpp
index 7fe4ab3..0ddd43e 100644
--- a/src/core/cpu/kernels/activation/NEON/fp16.cpp
+++ b/src/core/cpu/kernels/activation/NEON/fp16.cpp
@@ -26,7 +26,6 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Validate.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 
 #include <arm_neon.h>
 #include <cmath>
diff --git a/src/core/cpu/kernels/activation/NEON/fp32.cpp b/src/core/cpu/kernels/activation/NEON/fp32.cpp
index f1f2753..244ca57 100644
--- a/src/core/cpu/kernels/activation/NEON/fp32.cpp
+++ b/src/core/cpu/kernels/activation/NEON/fp32.cpp
@@ -26,7 +26,6 @@
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 
 #include <arm_neon.h>
 #include <cmath>
diff --git a/src/core/cpu/kernels/activation/NEON/qasymm8.cpp b/src/core/cpu/kernels/activation/NEON/qasymm8.cpp
index 7506a82..a121743 100644
--- a/src/core/cpu/kernels/activation/NEON/qasymm8.cpp
+++ b/src/core/cpu/kernels/activation/NEON/qasymm8.cpp
@@ -27,7 +27,6 @@
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 
 #include <arm_neon.h>
 #include <cmath>
diff --git a/src/core/cpu/kernels/activation/NEON/qasymm8_signed.cpp b/src/core/cpu/kernels/activation/NEON/qasymm8_signed.cpp
index 8f75abe..8b40bf8 100644
--- a/src/core/cpu/kernels/activation/NEON/qasymm8_signed.cpp
+++ b/src/core/cpu/kernels/activation/NEON/qasymm8_signed.cpp
@@ -26,7 +26,6 @@
 #include "src/core/NEON/NEAsymm.h"
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 
 #include <arm_neon.h>
 #include <cmath>
diff --git a/src/core/cpu/kernels/activation/NEON/qsymm16.cpp b/src/core/cpu/kernels/activation/NEON/qsymm16.cpp
index 9eee360..54b4182 100644
--- a/src/core/cpu/kernels/activation/NEON/qsymm16.cpp
+++ b/src/core/cpu/kernels/activation/NEON/qsymm16.cpp
@@ -28,7 +28,6 @@
 #include "src/core/NEON/NEMath.h"
 #include "src/core/NEON/NESymm.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 
 #include <arm_neon.h>
 #include <cmath>
diff --git a/src/core/cpu/kernels/activation/SVE/fp16.cpp b/src/core/cpu/kernels/activation/SVE/fp16.cpp
index 8208813..bf31fd7 100644
--- a/src/core/cpu/kernels/activation/SVE/fp16.cpp
+++ b/src/core/cpu/kernels/activation/SVE/fp16.cpp
@@ -24,7 +24,6 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/ITensorPack.h"
 #include "arm_compute/core/Window.h"
-#include "src/core/common/Validate.h"
 
 #include <cmath>
 #include <cstddef>
diff --git a/src/core/cpu/kernels/activation/SVE/fp32.cpp b/src/core/cpu/kernels/activation/SVE/fp32.cpp
index 55bdc99..75f9f8a 100644
--- a/src/core/cpu/kernels/activation/SVE/fp32.cpp
+++ b/src/core/cpu/kernels/activation/SVE/fp32.cpp
@@ -25,7 +25,6 @@
 #include "arm_compute/core/ITensorPack.h"
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/SVEMath.h"
-#include "src/core/common/Validate.h"
 
 #include <cmath>
 #include <cstddef>
diff --git a/src/core/cpu/kernels/activation/SVE/qasymm8.cpp b/src/core/cpu/kernels/activation/SVE/qasymm8.cpp
index 3d9476a..228b4ae 100644
--- a/src/core/cpu/kernels/activation/SVE/qasymm8.cpp
+++ b/src/core/cpu/kernels/activation/SVE/qasymm8.cpp
@@ -24,7 +24,6 @@
 
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Window.h"
-#include "src/core/common/Validate.h"
 
 #include <cmath>
 #include <cstddef>
diff --git a/src/core/cpu/kernels/activation/SVE/qasymm8_signed.cpp b/src/core/cpu/kernels/activation/SVE/qasymm8_signed.cpp
index 0b3d798..989f825 100644
--- a/src/core/cpu/kernels/activation/SVE/qasymm8_signed.cpp
+++ b/src/core/cpu/kernels/activation/SVE/qasymm8_signed.cpp
@@ -24,7 +24,6 @@
 #include "arm_compute/core/Helpers.h"
 #include "arm_compute/core/Window.h"
 #include "src/core/NEON/wrapper/wrapper.h"
-#include "src/core/common/Validate.h"
 
 #include <cmath>
 #include <cstddef>
diff --git a/src/core/cpu/kernels/activation/SVE/qsymm16.cpp b/src/core/cpu/kernels/activation/SVE/qsymm16.cpp
index dbaf267..6697487 100644
--- a/src/core/cpu/kernels/activation/SVE/qsymm16.cpp
+++ b/src/core/cpu/kernels/activation/SVE/qsymm16.cpp
@@ -25,7 +25,6 @@
 #include "arm_compute/core/ITensorPack.h"
 #include "arm_compute/core/Window.h"
 #include "arm_compute/core/experimental/Types.h"
-#include "src/core/common/Validate.h"
 
 #include <cmath>
 #include <cstddef>
diff --git a/src/core/cpu/kernels/floor/NEON/fp16.cpp b/src/core/cpu/kernels/floor/NEON/fp16.cpp
index 0d31eb7..f362676 100644
--- a/src/core/cpu/kernels/floor/NEON/fp16.cpp
+++ b/src/core/cpu/kernels/floor/NEON/fp16.cpp
@@ -23,8 +23,8 @@
  */
 #if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(ENABLE_FP16_KERNELS)
 
+#include "src/common/utils/Validate.h"
 #include "src/core/NEON/NEMath.h"
-#include "src/core/common/Validate.h"
 
 #include <arm_neon.h>
 #include <cmath>
diff --git a/src/core/cpu/kernels/floor/NEON/fp32.cpp b/src/core/cpu/kernels/floor/NEON/fp32.cpp
index dd63f9f..f5efb2e 100644
--- a/src/core/cpu/kernels/floor/NEON/fp32.cpp
+++ b/src/core/cpu/kernels/floor/NEON/fp32.cpp
@@ -21,8 +21,8 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
+#include "src/common/utils/Validate.h"
 #include "src/core/NEON/NEMath.h"
-#include "src/core/common/Validate.h"
 
 #include <arm_neon.h>
 #include <cmath>
diff --git a/src/core/gpu/cl/kernels/ClElementwiseKernel.cpp b/src/core/gpu/cl/kernels/ClElementwiseKernel.cpp
index 7d204b1..8f12eb2 100644
--- a/src/core/gpu/cl/kernels/ClElementwiseKernel.cpp
+++ b/src/core/gpu/cl/kernels/ClElementwiseKernel.cpp
@@ -25,8 +25,8 @@
 
 #include "arm_compute/core/CL/CLHelpers.h"
 #include "arm_compute/core/CL/ICLTensor.h"
+#include "src/common/utils/Validate.h"
 #include "src/core/CL/CLValidate.h"
-#include "src/core/common/Validate.h"
 #include "src/core/helpers/AutoConfiguration.h"
 #include "src/core/helpers/WindowHelpers.h"
 #include "support/Cast.h"