Reference picture resampling (RPR) based super-resolution with wavelet decomposition
AssigneeGuangdong Oppo Mobile Telecommunications Corp., Ltd.
InventorsCheolkon Jung, Hui Lan
Life of this patent
Estimating the term…
Abstract
A method for video processing includes receiving an input image via an initial feature extraction portion, where the initial feature portion includes a result of RPR up-sampling, a low resolution (LR) reconstructed frame, and an LR predicted frame; processing the input image by a first convolution layer; and processing the input image by a multi-level semantic information mapping portion to generate a combined image, where the multi-level semantic information mapping portion includes one or more residual spatial and channel attention blocks (RSCBs). Each of the RSCBs includes two or more residual spatial blocks (RSBs) and a channel attention (CA) block. Each of the RSBs includes a spatial attention (SA) block.
Claims19
A method for image processing, comprising:
- receiving an input image via an initial feature extraction portion, wherein the initial feature extraction portion includes a result of reference picture resampling (RPR) RPR up-sampling, a low resolution (LR) reconstructed picture, and an LR predicted picture,
- processing the input image by a first convolution layer;
- and
- processing the input image by a multi-level semantic information mapping portion to generate a combined image,
- wherein the multi-level semantic information mapping portion includes one or more residual spatial and channel attention blocks (RSCBs),
- wherein each of the RSCBs includes two or more residual spatial blocks (RSBs), and
- wherein processing the input image by the multi-level semantic information mapping portion includes:
- concatenating outputs of the RSBs;
- processing the concatenated outputs of the RSBs by a second convolutional layer so as to reduce a channel number;
- and
- obtaining a weight of channel by a channel attention (CA) block.
A system for video processing, the system comprising:
- a processor;
- and
- a memory configured to store instructions, when executed by the processor, to:
- receive an input image via an initial feature extraction portion, wherein the initial feature extraction portion includes a result of reference picture resampling (RPR) up-sampling, a low resolution (LR) reconstructed picture, and an LR predicted picture,
- process the input image by a first convolution layer;
- and
- process the input image by a multi-level semantic information mapping portion to generate a combined image, wherein the multi-level semantic information mapping portion includes one or more residual spatial and channel attention blocks (RSCBs), wherein each of the RSCBs includes two or more residual spatial blocks (RSBs)
- wherein processing the input image by the multi-level semantic information mapping portion includes:
- concatenating outputs of the RSBs;
- processing the concatenated outputs of the RSBs by a second convolutional layer so as to reduce a channel number;
- and
- obtaining a weight of channel by a channel attention (CA) block.
Dependencies are inferred from the claim text.