Adaptive Loop Filter Fusion Mode Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing adaptive loop filter (ALF) designs in video coding technologies generate filtering outputs independently, leading to inefficiencies in video coding processes, particularly in handling intermediate filtering results from different adaptation parameter sets (APS) or pre-defined filter sets.
Innovation Solution
The proposed solution involves a fusion mode that combines intermediate filtering results from various ALF processing units using a function with parameters derived from statistical, gradient, or other information, to generate a final filtering result, improving video coding efficiency by fusing outputs from multiple filters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If multiple intermediate filtering results are generated from different APS or pre-defined filter sets, then video coding performance is improved, but device complexity increases
Solution Approach 1:
The patent combines multiple intermediate filtering results from different APS or pre-defined filter sets into a single final filtering result through a fusion mode. This merging approach leverages the strengths of multiple filters to improve video coding performance while managing complexity through systematic integration rather than independent processing of each filter result
Solution Approach 2:
The fusion mode serves as a universal processing mechanism that can integrate filtering results from diverse sources including different APS and pre-defined filter sets. This multi-functional approach allows the system to handle various filter types and configurations through a single unified framework, improving versatility without proportionally increasing complexity
2Measurement precision
If fusion mode combines multiple intermediate filtering results, then filtering accuracy is improved, but processing time increases
Solution Approach 1:
The system performs preliminary filtering operations using multiple APS or pre-defined filter sets to generate intermediate filtering results before the fusion step. By pre-computing these intermediate results, the system prepares data in advance for the fusion operation, enabling more accurate final filtering while organizing processing in a structured sequence that can optimize time management
3Productivity
If conventional ALF generates filtering outputs independently, then processing simplicity is maintained, but video coding efficiency deteriorates
Solution Approach 1:
The patent merges multiple independent filtering outputs into a unified final result through fusion mode, transforming the conventional independent processing structure into an integrated system. This merging improves video coding efficiency by leveraging complementary information from different filters while maintaining a manageable processing structure through systematic fusion operations
Data Source
AI summary
A method of processing media data. The method includes fusing a plurality of intermediate filtering results of an adaptive loop filter (ALF) processing unit with a fusion mode to generate a final filtering result; and performing a conversion between a video including the ALF processing unit and a bitstream of the video based on the final filtering result. A corresponding video coding apparatus and non-transitory computer-readable recording medium are also disclosed.