diff --git a/src/runtime/CL/functions/CLDirectDeconvolutionLayer.cpp b/src/runtime/CL/functions/CLDirectDeconvolutionLayer.cpp
index c1a39ef..b8089d8 100644
--- a/src/runtime/CL/functions/CLDirectDeconvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLDirectDeconvolutionLayer.cpp
@@ -63,13 +63,8 @@
 
     ARM_COMPUTE_RETURN_ERROR_ON(weights->dimension(idx_w) != weights->dimension(idx_h));
     ARM_COMPUTE_RETURN_ERROR_ON(weights->dimension(idx_w) < 1);
-    ARM_COMPUTE_RETURN_ERROR_ON(!info.padding_is_symmetric());
 
-    const unsigned int stride_x = info.stride().first;
-    const unsigned int stride_y = info.stride().second;
-
-    auto out_dims = deconvolution_output_dimensions(input->dimension(idx_w), input->dimension(idx_h), weights->dimension(idx_w), weights->dimension(idx_h),
-                                                    info.pad().first, info.pad().second, stride_x, stride_y);
+    auto out_dims = deconvolution_output_dimensions(input->dimension(idx_w), input->dimension(idx_h), weights->dimension(idx_w), weights->dimension(idx_h), info);
 
     const TensorShape output_shape = compute_deconvolution_output_shape(out_dims, *input, *weights);
 
@@ -92,9 +87,11 @@
     ARM_COMPUTE_RETURN_ERROR_ON_MSG(output->dimension(idx_h) != output_shape[idx_h], "Output's height is invalid.");
     ARM_COMPUTE_RETURN_ERROR_ON_MSG(output->dimension(idx_c) != output_shape[idx_c], "Output's depth is invalid.");
 
-    unsigned int        padx            = 0;
-    unsigned int        pady            = 0;
-    const TensorShape   scale_out_shape = compute_deconvolution_upsampled_shape(*input, *weights, stride_x, stride_y, out_dims, padx, pady);
+    unsigned int        deconv_pad_x = 0;
+    unsigned int        deconv_pad_y = 0;
+    const unsigned int  stride_x = info.stride().first;
+    const unsigned int  stride_y = info.stride().second;
+    const TensorShape   scale_out_shape = compute_deconvolution_upsampled_shape(*input, *weights, stride_x, stride_y, out_dims, deconv_pad_x, deconv_pad_y);
     TensorInfo          scale_out_info(input->clone()->set_is_resizable(true).reset_padding().set_tensor_shape(scale_out_shape).set_data_layout(data_layout));
     const PadStrideInfo conv_info(1, 1, 0, 0, 0, 0, DimensionRoundingType::CEIL);
 
@@ -109,6 +106,10 @@
 {
     ARM_COMPUTE_ERROR_ON_NULLPTR(input, weights, output);
 
+    const unsigned int pad_left   = info.pad_left();
+    const unsigned int pad_right  = info.pad_right();
+    const unsigned int pad_top    = info.pad_top();
+    const unsigned int pad_bottom = info.pad_bottom();
     const unsigned int stride_x = info.stride().first;
     const unsigned int stride_y = info.stride().second;
 
@@ -122,8 +123,7 @@
     _weights_flipped.allocator()->init(weights->info()->clone()->set_data_layout(data_layout));
     _flip_weights.configure(weights, &_weights_flipped, &_flip_axis);
 
-    auto out_dims = deconvolution_output_dimensions(input->info()->dimension(idx_w), input->info()->dimension(idx_h), weights->info()->dimension(idx_w), weights->info()->dimension(idx_h),
-                                                    info.pad().first, info.pad().second, stride_x, stride_y);
+    auto out_dims = deconvolution_output_dimensions(input->info()->dimension(idx_w), input->info()->dimension(idx_h), weights->info()->dimension(idx_w), weights->info()->dimension(idx_h), info);
 
     const TensorShape output_shape = compute_deconvolution_output_shape(out_dims, *input->info(), *weights->info());
 
@@ -138,16 +138,30 @@
     _memory_group.manage(&_scaled_output);
 
     // Find the upsampled dimensions and the padding needed for the convolution with stride 1 in order to match output shape
-    unsigned int      padx            = 0;
-    unsigned int      pady            = 0;
-    const TensorShape scale_out_shape = compute_deconvolution_upsampled_shape(*input->info(), *weights->info(), stride_x, stride_y, out_dims, padx, pady);
+    unsigned int      deconv_pad_x    = 0;
+    unsigned int      deconv_pad_y    = 0;
+    const TensorShape scale_out_shape = compute_deconvolution_upsampled_shape(*input->info(), *weights->info(), stride_x, stride_y, out_dims, deconv_pad_x, deconv_pad_y);
+
+    unsigned int deconv_pad_left  = pad_right > pad_left ? pad_right - pad_left : 0;
+    unsigned int deconv_pad_right = pad_left > pad_right ? pad_left - pad_right : 0;
+    deconv_pad_x -= deconv_pad_left + deconv_pad_right;
+    ARM_COMPUTE_ERROR_ON((deconv_pad_x % 2) != 0);
+    deconv_pad_left  += deconv_pad_x / 2;
+    deconv_pad_right += deconv_pad_x / 2;
+
+    unsigned int deconv_pad_top    = pad_bottom > pad_top ? pad_bottom - pad_top : 0;
+    unsigned int deconv_pad_bottom = pad_top > pad_bottom ? pad_top - pad_bottom : 0;
+    deconv_pad_y -= deconv_pad_top + deconv_pad_bottom;
+    ARM_COMPUTE_ERROR_ON((deconv_pad_y % 2) != 0);
+    deconv_pad_top    += deconv_pad_y / 2;
+    deconv_pad_bottom += deconv_pad_y / 2;
 
     TensorInfo scale_out_info(scale_out_shape, 1, input->info()->data_type(), input->info()->quantization_info());
     scale_out_info.set_data_layout(data_layout);
     _scaled_output.allocator()->init(scale_out_info);
 
     // configure scale function
-    const PadStrideInfo upsample_info(stride_x, stride_y, padx / 2, pady / 2);
+    const PadStrideInfo upsample_info(stride_x, stride_y, deconv_pad_left, deconv_pad_right, deconv_pad_top, deconv_pad_bottom, DimensionRoundingType::FLOOR);
     _scale_f.configure(input, &_scaled_output, upsample_info);
 
     // Setup the function to convolve the upscaled output
diff --git a/src/runtime/CL/functions/CLGEMMDeconvolutionLayer.cpp b/src/runtime/CL/functions/CLGEMMDeconvolutionLayer.cpp
index 36a120e..78e1ae7 100644
--- a/src/runtime/CL/functions/CLGEMMDeconvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLGEMMDeconvolutionLayer.cpp
@@ -151,8 +151,8 @@
         ARM_COMPUTE_RETURN_ON_ERROR(CLGEMM::validate(&input->clone()->set_tensor_shape(nhwc_input_shape).set_is_resizable(true), &reshaped_t_info, nullptr, &gemm_output_info, 1.0f, 0.0f, gemm_info));
     }
 
-    auto out_dims = deconvolution_output_dimensions(input->dimension(idx_w), input->dimension(idx_h), weights->dimension(idx_w), weights->dimension(idx_h),
-                                                    0, 0, deconv_info.stride().first, deconv_info.stride().second);
+    const PadStrideInfo stride_info(deconv_info.stride().first, deconv_info.stride().second);
+    auto out_dims = deconvolution_output_dimensions(input->dimension(idx_w), input->dimension(idx_h), weights->dimension(idx_w), weights->dimension(idx_h), stride_info);
     const TensorShape deconv_shape       = misc::shape_calculator::compute_deconvolution_output_shape(out_dims, *input, *weights);
     TensorInfo        col2im_output_info = gemm_output_info.clone()->set_tensor_shape(deconv_shape).set_is_resizable(true);