diff --git a/src/core/CL/CLKernelLibrary.cpp b/src/core/CL/CLKernelLibrary.cpp
index 8a309ec..1bc1036 100644
--- a/src/core/CL/CLKernelLibrary.cpp
+++ b/src/core/CL/CLKernelLibrary.cpp
@@ -316,6 +316,8 @@
     { "normalization_layer_in_map", "normalization_layer.cl" },
     { "normalize_planar_yuv_layer_nchw", "normalize_planar_yuv_layer.cl" },
     { "normalize_planar_yuv_layer_nhwc", "normalize_planar_yuv_layer.cl" },
+    { "normalize_planar_yuv_layer_q8_nchw", "normalize_planar_yuv_layer_quantized.cl" },
+    { "normalize_planar_yuv_layer_q8_nhwc", "normalize_planar_yuv_layer_quantized.cl" },
     { "NV12_to_IYUV_bt709", "color_convert.cl" },
     { "NV12_to_RGB888_bt709", "color_convert.cl" },
     { "NV12_to_RGBA8888_bt709", "color_convert.cl" },
@@ -698,6 +700,10 @@
 #include "./cl_kernels/normalize_planar_yuv_layer.clembed"
     },
     {
+        "normalize_planar_yuv_layer_quantized.cl",
+#include "./cl_kernels/normalize_planar_yuv_layer_quantized.clembed"
+    },
+    {
         "batchnormalization_layer.cl",
 #include "./cl_kernels/batchnormalization_layer.clembed"
     },
diff --git a/src/core/CL/cl_kernels/normalize_planar_yuv_layer.cl b/src/core/CL/cl_kernels/normalize_planar_yuv_layer.cl
index dc66524..a105968 100644
--- a/src/core/CL/cl_kernels/normalize_planar_yuv_layer.cl
+++ b/src/core/CL/cl_kernels/normalize_planar_yuv_layer.cl
@@ -27,7 +27,7 @@
 
 #define TYPE VEC_DATA_TYPE(DATA_TYPE, VEC_SIZE)
 
-/** Apply normalize_planar_yuv layer on tensors with NCHW format.
+/** Apply normalize_planar_yuv layer on tensors with NCHW data layout.
  *
  * @note Data type should be given as a preprocessor argument using -DDATA_TYPE=type. e.g. -DDATA_TYPE=float
  * @note Vector size should be given as a preprocessor argument using -DVEC_SIZE e.g. -DVEC_SIZE=8
@@ -70,8 +70,8 @@
 
     const uint current_slice = get_global_id(2) % NUM_CHANNELS;
 
-    const DATA_TYPE curr_mean = *((__global DATA_TYPE *)(mean.ptr + current_slice * mean.stride_x));
-    const DATA_TYPE curr_std  = *((__global DATA_TYPE *)(std.ptr + current_slice * std.stride_x));
+    const DATA_TYPE curr_mean = *((__global DATA_TYPE *)(mean.ptr + current_slice * sizeof(DATA_TYPE)));
+    const DATA_TYPE curr_std  = *((__global DATA_TYPE *)(std.ptr + current_slice * sizeof(DATA_TYPE)));
 
     TYPE data = VLOAD(VEC_SIZE)(0, (__global DATA_TYPE *)src.ptr);
     TYPE res  = (data - curr_mean) / curr_std;
@@ -80,7 +80,7 @@
     (res, 0, (__global DATA_TYPE *)dst.ptr);
 }
 
-/** Apply normalize_planar_yuv layer on tensors with NHWC format.
+/** Apply normalize_planar_yuv layer on tensors with NHWC data layout.
  *
  * @note Data type should be given as a preprocessor argument using -DDATA_TYPE=type. e.g. -DDATA_TYPE=float
  * @note Vector size should be given as a preprocessor argument using -DVEC_SIZE e.g. -DVEC_SIZE=8
@@ -122,8 +122,8 @@
 
     const uint current_slice = get_global_id(0);
 
-    const TYPE curr_mean = VLOAD(VEC_SIZE)(0, (__global DATA_TYPE *)(mean.ptr + current_slice * VEC_SIZE * mean.stride_x));
-    const TYPE curr_std  = VLOAD(VEC_SIZE)(0, (__global DATA_TYPE *)(std.ptr + current_slice * VEC_SIZE * std.stride_x));
+    const TYPE curr_mean = VLOAD(VEC_SIZE)(0, (__global DATA_TYPE *)(mean.ptr + current_slice * VEC_SIZE * sizeof(DATA_TYPE)));
+    const TYPE curr_std  = VLOAD(VEC_SIZE)(0, (__global DATA_TYPE *)(std.ptr + current_slice * VEC_SIZE * sizeof(DATA_TYPE)));
 
     TYPE data = VLOAD(VEC_SIZE)(0, (__global DATA_TYPE *)src.ptr);
     TYPE res  = (data - curr_mean) / curr_std;
diff --git a/src/core/CL/cl_kernels/normalize_planar_yuv_layer_quantized.cl b/src/core/CL/cl_kernels/normalize_planar_yuv_layer_quantized.cl
new file mode 100644
index 0000000..925975d
--- /dev/null
+++ b/src/core/CL/cl_kernels/normalize_planar_yuv_layer_quantized.cl
@@ -0,0 +1,158 @@
+/*
+ * Copyright (c) 2018 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 "helpers.h"
+
+#if defined(DATA_TYPE) && defined(VEC_SIZE) && defined(OFFSET) && defined(SCALE)
+
+#define TYPE VEC_DATA_TYPE(DATA_TYPE, VEC_SIZE)
+#define OFFSET_FLT ((float)OFFSET)
+#define SCALE_FLT ((float)SCALE)
+
+#if defined(NUM_CHANNELS)
+
+/** Apply normalize_planar_yuv layer on tensors with NCHW data layout.
+ *
+ * @note Data type should be given as a preprocessor argument using -DDATA_TYPE=type. e.g. -DDATA_TYPE=float
+ * @note Vector size should be given as a preprocessor argument using -DVEC_SIZE e.g. -DVEC_SIZE=8
+ * @note The depth of the input tensor should be given as a preprocessor argument using -DNUM_CHANNELS e.g. -DNUM_CHANNELS=8
+ * @note The quantization offset should be given as a preprocessor argument using -DOFFSET e.g. -DOFFSET=8
+ * @note The quantization scale should be given as a preprocessor argument using -DSCALE e.g. -DSCALE=8
+ *
+ * @param[in]  src_ptr                            Pointer to the first source tensor. Supported data types: QASYMM8
+ * @param[in]  src_stride_x                       Stride of the first source tensor in X dimension (in bytes)
+ * @param[in]  src_step_x                         input_stride_x * number of elements along X processed per workitem(in bytes)
+ * @param[in]  src_stride_y                       Stride of the first source tensor in Y dimension (in bytes)
+ * @param[in]  src_step_y                         input_stride_y * number of elements along Y processed per workitem(in bytes)
+ * @param[in]  src_stride_z                       Stride of the first source tensor in Z dimension (in bytes)
+ * @param[in]  src_step_z                         input_stride_z * number of elements along Z processed per workitem(in bytes)
+ * @param[in]  src_offset_first_element_in_bytes  The offset of the first element in the first source tensor
+ * @param[out] dst_ptr                            Pointer to the destination tensor. Supported data types: same as @p src_ptr
+ * @param[in]  dst_stride_x                       Stride of the destination tensor in X dimension (in bytes)
+ * @param[in]  dst_step_x                         output_stride_x * number of elements along X processed per workitem(in bytes)
+ * @param[in]  dst_stride_y                       Stride of the destination tensor in Y dimension (in bytes)
+ * @param[in]  dst_step_y                         output_stride_y * number of elements along Y processed per workitem(in bytes)
+ * @param[in]  dst_stride_z                       Stride of the destination tensor in Z dimension (in bytes)
+ * @param[in]  dst_step_z                         output_stride_z * number of elements along Z processed per workitem(in bytes)
+ * @param[in]  dst_offset_first_element_in_bytes  The offset of the first element in the destination tensor
+ * @param[in]  mean_ptr                           Pointer to the mean source tensor. Supported data types: same as @p src_ptr
+ * @param[in]  mean_stride_x                      Stride of the mean source tensor in X dimension (in bytes)
+ * @param[in]  mean_step_x                        mean_stride_x * number of elements along X processed per workitem(in bytes)
+ * @param[in]  mean_offset_first_element_in_bytes The offset of the first element in the mean source tensor
+ * @param[in]  std_ptr                            Pointer to the std tensor. Supported data types: same as @p src_ptr
+ * @param[in]  std_stride_x                       Stride of the std tensor in X dimension (in bytes)
+ * @param[in]  std_step_x                         std_stride_x * number of elements along X processed per workitem(in bytes)
+ * @param[in]  std_offset_first_element_in_bytes  The offset of the first element in the var source tensor
+ */
+__kernel void normalize_planar_yuv_layer_q8_nchw(TENSOR3D_DECLARATION(src),
+                                                 TENSOR3D_DECLARATION(dst),
+                                                 VECTOR_DECLARATION(mean),
+                                                 VECTOR_DECLARATION(std))
+{
+    Tensor3D src  = CONVERT_TO_TENSOR3D_STRUCT(src);
+    Tensor3D dst  = CONVERT_TO_TENSOR3D_STRUCT(dst);
+    Vector   mean = CONVERT_TO_VECTOR_STRUCT(mean);
+    Vector   std  = CONVERT_TO_VECTOR_STRUCT(std);
+
+    const uint current_slice = get_global_id(2) % NUM_CHANNELS;
+
+    float16 curr_mean_flt = (float16)(*((__global DATA_TYPE *)(mean.ptr + current_slice * sizeof(DATA_TYPE))));
+    curr_mean_flt         = round(curr_mean_flt - OFFSET_FLT) * SCALE_FLT;
+
+    float16 curr_std_flt = (float16)(*((__global DATA_TYPE *)(std.ptr + current_slice * sizeof(DATA_TYPE))));
+    curr_std_flt         = round(curr_std_flt - OFFSET_FLT) * SCALE_FLT;
+
+    float16 data_flt = CONVERT(VLOAD(VEC_SIZE)(0, (__global DATA_TYPE *)src.ptr), float16);
+    data_flt         = round(data_flt - OFFSET_FLT) * SCALE_FLT;
+
+    // Perform normalization
+    float16 res_flt = (data_flt - curr_mean_flt) / curr_std_flt;
+
+    const TYPE res_u8 = CONVERT_SAT(round(res_flt / SCALE_FLT) + OFFSET_FLT, TYPE);
+    VSTORE(VEC_SIZE)
+    (res_u8, 0, (__global DATA_TYPE *)dst.ptr);
+}
+
+#endif // defined(NUM_CHANNELS)
+
+/** Apply normalize_planar_yuv layer on tensors with NHWC data layout.
+ *
+ * @note Data type should be given as a preprocessor argument using -DDATA_TYPE=type. e.g. -DDATA_TYPE=float
+ * @note Vector size should be given as a preprocessor argument using -DVEC_SIZE e.g. -DVEC_SIZE=8
+ * @note The quantization offset should be given as a preprocessor argument using -DOFFSET e.g. -DOFFSET=8
+ * @note The quantization scale should be given as a preprocessor argument using -DSCALE e.g. -DSCALE=8
+ *
+ * @param[in]  src_ptr                            Pointer to the first source tensor. Supported data types: QASYMM8
+ * @param[in]  src_stride_x                       Stride of the first source tensor in X dimension (in bytes)
+ * @param[in]  src_step_x                         input_stride_x * number of elements along X processed per workitem(in bytes)
+ * @param[in]  src_stride_y                       Stride of the first source tensor in Y dimension (in bytes)
+ * @param[in]  src_step_y                         input_stride_y * number of elements along Y processed per workitem(in bytes)
+ * @param[in]  src_stride_z                       Stride of the first source tensor in Z dimension (in bytes)
+ * @param[in]  src_step_z                         input_stride_z * number of elements along Z processed per workitem(in bytes)
+ * @param[in]  src_offset_first_element_in_bytes  The offset of the first element in the first source tensor
+ * @param[out] dst_ptr                            Pointer to the destination tensor. Supported data types: same as @p src_ptr
+ * @param[in]  dst_stride_x                       Stride of the destination tensor in X dimension (in bytes)
+ * @param[in]  dst_step_x                         output_stride_x * number of elements along X processed per workitem(in bytes)
+ * @param[in]  dst_stride_y                       Stride of the destination tensor in Y dimension (in bytes)
+ * @param[in]  dst_step_y                         output_stride_y * number of elements along Y processed per workitem(in bytes)
+ * @param[in]  dst_stride_z                       Stride of the destination tensor in Z dimension (in bytes)
+ * @param[in]  dst_step_z                         output_stride_z * number of elements along Z processed per workitem(in bytes)
+ * @param[in]  dst_offset_first_element_in_bytes  The offset of the first element in the destination tensor
+ * @param[in]  mean_ptr                           Pointer to the mean source tensor. Supported data types: same as @p src_ptr
+ * @param[in]  mean_stride_x                      Stride of the mean source tensor in X dimension (in bytes)
+ * @param[in]  mean_step_x                        mean_stride_x * number of elements along X processed per workitem(in bytes)
+ * @param[in]  mean_offset_first_element_in_bytes The offset of the first element in the mean source tensor
+ * @param[in]  std_ptr                            Pointer to the std tensor. Supported data types: same as @p src_ptr
+ * @param[in]  std_stride_x                       Stride of the std tensor in X dimension (in bytes)
+ * @param[in]  std_step_x                         std_stride_x * number of elements along X processed per workitem(in bytes)
+ * @param[in]  std_offset_first_element_in_bytes  The offset of the first element in the var source tensor
+ */
+__kernel void normalize_planar_yuv_layer_q8_nhwc(TENSOR3D_DECLARATION(src),
+                                                 TENSOR3D_DECLARATION(dst),
+                                                 VECTOR_DECLARATION(mean),
+                                                 VECTOR_DECLARATION(std))
+{
+    Tensor3D src  = CONVERT_TO_TENSOR3D_STRUCT(src);
+    Tensor3D dst  = CONVERT_TO_TENSOR3D_STRUCT(dst);
+    Vector   mean = CONVERT_TO_VECTOR_STRUCT(mean);
+    Vector   std  = CONVERT_TO_VECTOR_STRUCT(std);
+
+    const uint current_slice = get_global_id(0);
+
+    float16 curr_mean_flt = CONVERT(VLOAD(VEC_SIZE)(0, (__global DATA_TYPE *)(mean.ptr + current_slice * VEC_SIZE * sizeof(DATA_TYPE))), float16);
+    curr_mean_flt         = round(curr_mean_flt - OFFSET_FLT) * SCALE_FLT;
+
+    float16 curr_std_flt = CONVERT(VLOAD(VEC_SIZE)(0, (__global DATA_TYPE *)(std.ptr + current_slice * VEC_SIZE * sizeof(DATA_TYPE))), float16);
+    curr_std_flt         = round(curr_std_flt - OFFSET_FLT) * SCALE_FLT;
+
+    float16 data_flt = CONVERT(VLOAD(VEC_SIZE)(0, (__global DATA_TYPE *)src.ptr), float16);
+    data_flt         = round(data_flt - OFFSET_FLT) * (SCALE_FLT);
+
+    // Perform normalization
+    float16 res_flt = (data_flt - curr_mean_flt) / curr_std_flt;
+
+    const TYPE res_u8 = CONVERT_SAT(round(res_flt / SCALE_FLT) + OFFSET_FLT, TYPE);
+    VSTORE(VEC_SIZE)
+    (res_u8, 0, (__global DATA_TYPE *)dst.ptr);
+}
+#endif // defined(DATA_TYPE) && defined(VEC_SIZE) && defined(OFFSET) && defined(SCALE)
diff --git a/src/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.cpp b/src/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.cpp
index 31451ef..a44507b 100644
--- a/src/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.cpp
+++ b/src/core/CL/kernels/CLNormalizePlanarYUVLayerKernel.cpp
@@ -42,7 +42,7 @@
 Status validate_arguments(const ITensorInfo *input, const ITensorInfo *output, const ITensorInfo *mean, const ITensorInfo *std)
 {
     ARM_COMPUTE_RETURN_ERROR_ON_F16_UNSUPPORTED(input);
-    ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::F16, DataType::F32);
+    ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::QASYMM8, DataType::F16, DataType::F32);
     ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(output);
 
     ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(input, mean, std);
@@ -111,15 +111,25 @@
 
     const unsigned int num_elems_processed_per_iteration = 16 / input->info()->element_size();
     const unsigned int channel_idx                       = get_data_layout_dimension_index(input->info()->data_layout(), DataLayoutDimension::CHANNEL);
+    const DataType     dt                                = input->info()->data_type();
 
     // Set build options
     CLBuildOptions build_opts;
-    build_opts.add_option(("-DDATA_TYPE=" + get_cl_type_from_data_type(input->info()->data_type())));
+    build_opts.add_option(("-DDATA_TYPE=" + get_cl_type_from_data_type(dt)));
     build_opts.add_option(("-DVEC_SIZE=" + support::cpp11::to_string(num_elems_processed_per_iteration)));
     build_opts.add_option(("-DNUM_CHANNELS=" + support::cpp11::to_string(input->info()->dimension(channel_idx))));
 
+    std::string kernel_name = "normalize_planar_yuv_layer_";
+    if(is_data_type_quantized(dt))
+    {
+        build_opts.add_option(("-DOFFSET=" + support::cpp11::to_string(input->info()->quantization_info().offset)));
+        build_opts.add_option(("-DSCALE=" + support::cpp11::to_string(input->info()->quantization_info().scale)));
+        kernel_name += "q8_";
+    }
+
     // Create kernel
-    _kernel = static_cast<cl::Kernel>(CLKernelLibrary::get().create_kernel("normalize_planar_yuv_layer_" + lower_string(string_from_data_layout(input->info()->data_layout())), build_opts.options()));
+    kernel_name += lower_string(string_from_data_layout(input->info()->data_layout()));
+    _kernel = static_cast<cl::Kernel>(CLKernelLibrary::get().create_kernel(kernel_name, build_opts.options()));
 
     // Configure kernel window
     auto win_config = validate_and_configure_window(input->info(), output->info(), mean->info(), std->info());
@@ -130,7 +140,7 @@
     _config_id = "normalize_planar_yuv_layer_";
     _config_id += lower_string(string_from_data_layout(input->info()->data_layout()));
     _config_id += "_";
-    _config_id += lower_string(string_from_data_type(input->info()->data_type()));
+    _config_id += lower_string(string_from_data_type(dt));
     _config_id += "_";
     _config_id += support::cpp11::to_string(input->info()->dimension(0));
     _config_id += "_";