Cross-Component Adaptive Loop Filtering for Chroma Block Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cross-component adaptive loop filtering (CC-ALF) designs suffer from low classification accuracy, leading to poor performance due to rough adaptive selection at the CTU level, which affects the accuracy of content classification and reduces the adaptive capability of CC-ALF, thereby impacting coding and decoding efficiency.
Innovation Solution
Enhance CC-ALF by adding a block classification category for the chroma component, utilizing the classification information of the luma component to improve content classification accuracy and adaptive capability, thereby enhancing filtering performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If classification-based filter coefficient selection is used in current CC-ALF design, then the filtering process can be adapted to different video content characteristics, but the classification accuracy is low which causes poor CC-ALF performance
Solution Approach 1:
The patent transitions from traditional classification-based filter selection to a regression-based continuous parameter approach. Instead of categorizing video blocks into discrete classes and selecting filters accordingly, the invention uses regression to directly predict optimal filter coefficients as continuous values, thereby eliminating the accuracy limitations of classification while maintaining adaptability to different video content characteristics.
Solution Approach 2:
The invention changes the parameter representation from discrete classification labels to continuous regression outputs. By predicting filter coefficients as continuous parameters rather than selecting from predefined categories, the system achieves higher precision in adapting to video content characteristics, directly resolving the low classification accuracy problem while improving overall CC-ALF performance.
2Productivity
If traditional CC-ALF processing is used, then the decoding process can be maintained at current complexity levels, but the encoding and decoding efficiency remains suboptimal due to poor classification accuracy
Solution Approach 1:
The patent replaces the mechanical classification system with a regression-based prediction system. Instead of using discrete classification rules and threshold-based filter selection, the invention employs regression models to directly predict optimal filter coefficients, thereby improving content analysis precision and subsequently enhancing encoding and decoding efficiency through more accurate adaptive filtering.
Solution Approach 2:
By changing from discrete classification parameters to continuous regression parameters, the system achieves superior content characterization accuracy. This parameter transformation enables more precise adaptation to video content, which directly improves filtering performance and overall processing efficiency in both encoding and decoding stages.
Data Source
AI summary
A loop filtering method, a video coding and decoding method and apparatus, a medium, and an electronic device that can obtain block classification information of a luma component in a video image frame in a case that adaptive loop filtering is performed (S710), use the block the block classification information to determine block classification information of a chroma component in the video image frame in a case that cross-component adaptive loop filtering is performed (S720); and select a corresponding filter coefficient to perform cross-component adaptive loop filtering on the chroma component according to the block classification information of the chroma component.