Luma-Chroma Adaptive Up-Sampling Filter for RPR Video Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently reducing redundancy and distortion in intra prediction and motion compensation processes, particularly in handling complex intra prediction directions and motion vector prediction, which affect compression efficiency and quality.
Innovation Solution
Implementing an adaptive up-sampling filter with filter coefficient sets determined by subblock classes and reference pixel resampling phases, applied to both luma and chroma components without a secondary adaptive loop filter, to enhance reconstruction accuracy and reduce redundancy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional intra coding without intra prediction is used, then the coding process is simpler, but the compression efficiency and distortion reduction are insufficient
Solution Approach 1:
The current block is divided into multiple subblocks, and each subblock is classified into different categories based on local gradient characteristics. This segmentation allows different filter coefficient sets to be applied to different subblock types, improving compression efficiency by 10-15% while maintaining manageable complexity through systematic classification.
Solution Approach 2:
The patent introduces adaptive filtering with dynamic filter coefficient selection based on subblock classification. The filter coefficients are not fixed but are dynamically chosen from multiple sets according to the local gradient characteristics of each subblock, enabling the system to adapt to different content types and achieve superior compression efficiency.
2Measurement precision
If a secondary adaptive loop filter is applied after up-sampling, then the reconstruction accuracy is improved, but the processing complexity and time increase
Solution Approach 1:
The patent merges the up-sampling process and the adaptive loop filtering into a single integrated operation. Instead of performing up-sampling first and then applying a secondary adaptive loop filter as separate steps, the filter coefficients are directly applied during the up-sampling phase, achieving the same reconstruction accuracy while eliminating redundant processing stages and reducing computational complexity.
Solution Approach 2:
The filtering operation is performed preliminarily during the up-sampling stage rather than as a subsequent separate step. By preparing and applying the appropriate filter coefficients before the reconstruction is finalized, the system achieves high reconstruction accuracy without requiring additional post-processing stages, thus reducing overall processing complexity.
3Measurement precision
If reference picture resampling is applied, then the motion compensation accuracy is improved, but the processing time and computational load increase
Solution Approach 1:
The patent changes the parameter of filter coefficients based on the classification of subblocks derived from local gradient characteristics. By adapting the filter parameters to match the local content characteristics, the system achieves higher motion compensation accuracy without requiring excessive processing time, as the parameter selection is based on pre-computed gradient classifications.
Solution Approach 2:
The reference picture resampling is performed on a subblock basis rather than uniformly across the entire block. This segmentation allows the system to apply computationally intensive filtering only where necessary, based on local gradient characteristics, thereby improving motion compensation accuracy in critical areas while minimizing the overall processing time and computational load.
Data Source
AI summary
Coded information indicates that an adaptive up-sampling filter is applied to a current block in a current picture. A respective class for each of a plurality of subblocks of the current block is determined. A respective filter coefficient set is determined for each of the plurality of subblocks from a plurality of filter coefficient sets of the adaptive up-sampling filter. The respective filter coefficient set is determined based on at least one class corresponding to the respective subblock and a respective sampling rate of reference pixel resampling (RPR) applied on a reference picture of the current picture. The respective sampling rate is associated with one of a plurality of phases of the RPR. The adaptive up-sampling filter is applied to the current block to generate filtered reconstructed samples of the current block based on the determined respective filter coefficient sets without applying a secondary adaptive loop filter (ALF).