Motion-Compensated Picture Padding for VVC Boundary Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards face inefficiencies in handling out-of-boundary pixel padding during the reconstruction stage, particularly in versatile video coding (VVC), where conventional methods provide limited information and suboptimal approximations, leading to impaired video quality.
Innovation Solution
A method and apparatus for generating padded samples outside the picture boundary using motion-compensated reconstruction, incorporating multi-directional boundary padding (MDBP) and motion-compensated boundary pixel padding techniques, which utilize angular intra-prediction and motion compensation to extend the reference frame border, optimizing pixel prediction and reducing signaling costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional boundary padding methods are used in VVC, then the implementation is simple, but the video quality deteriorates due to limited information and suboptimal approximations
Solution Approach 1:
The patent applies preliminary action by generating motion-compensated boundary padding samples before the actual reconstruction process. The decoder generates padded samples for reference pictures using motion compensation techniques in advance, ensuring that boundary regions have sufficient information before being used for prediction, thereby improving video quality without requiring complex post-processing
Solution Approach 2:
The patent uses motion compensation as an intermediary mechanism to generate boundary padding samples. Instead of directly copying or extrapolating boundary pixels, the system uses motion vectors and reference picture data as intermediaries to compute more accurate padded samples, resolving the contradiction between simplicity and quality
2Productivity
If motion-compensated boundary padding is implemented, then coding efficiency improves, but computational complexity increases
Solution Approach 1:
The patent applies local quality by differentiating the padding approach based on location. Motion-compensated padding is applied specifically to boundary regions where it provides the most benefit, while interior regions use standard reconstruction methods. This localized application improves coding efficiency at boundary areas without unnecessarily increasing complexity across the entire processing system
Solution Approach 2:
The patent changes parameters by utilizing motion vectors and reference picture indices that are already available in the video coding system. By reusing existing parameters rather than introducing new complex computations, the system achieves improved coding efficiency with minimal additional computational burden
3Measurement precision
If extended motion-compensated reconstructed blocks are generated, then padding accuracy improves, but processing time increases
Solution Approach 1:
The patent performs preliminary action by generating extended motion-compensated blocks that include padding regions during the initial reconstruction phase. This preliminary generation of accurate padded samples prevents the need for time-consuming post-processing or iterative refinement, achieving both high accuracy and efficient processing
Solution Approach 2:
The patent merges the padding generation process with the main motion compensation and reconstruction operations. Instead of treating padding as a separate post-processing step, the system combines boundary sample generation with the primary reconstruction workflow, thereby improving padding accuracy without adding significant processing time
Data Source
AI summary
A method and apparatus for padding out-of-boundary pixels are disclosed. According to the method, input data associated with a current block located at or near a picture boundary are received, wherein the input data comprise prediction data and reconstructed residual data related to the current block. An extended motion-compensated reconstructed block for the current block is generated based on the prediction data and the reconstructed residual data, wherein the extended motion-compensated reconstructed block for the current block is inter coded and comprises a padded area located outside the picture boundary and a reconstructed current block. At least one in-loop filter is applied to the extended motion-compensated reconstructed block.