Auxiliary-Aware Neural Network Filtering for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies, such as MPEG-2, MPEG-4, ITU-T.263, ITU-T.264/MPEG-4 AVC, ITU-T.265 HEVC, and VVC, require improvements in coding efficiency and coding effectiveness, particularly in handling different interpretations of visual quality improvement and auxiliary inputs for neural-network post-processing filters.
Innovation Solution
A method and apparatus for video processing that applies a neural network filter to video units based on auxiliary information like prediction, partitioning, and coding information, and generates bitstreams, enhancing coding efficiency and effectiveness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If neural network filter is applied to video units using conventional auxiliary information, then visual quality improvement is achieved, but coding efficiency and coding effectiveness are insufficient
Solution Approach 1:
The patent changes the parameters of auxiliary information by introducing prediction information, partitioning information, and coding information from previously coded video units. These parameter changes enable the neural network filter to process video data more efficiently, achieving both improved visual quality and enhanced coding efficiency simultaneously
Solution Approach 2:
The patent applies preliminary action by utilizing prediction information and coding information from previously coded video units before processing the current video unit. This allows the neural network filter to leverage already-processed data, reducing computational redundancy and improving overall coding efficiency while maintaining visual quality
2Productivity
If more auxiliary information is provided to neural network filter, then coding effectiveness is improved, but device complexity increases
Solution Approach 1:
The patent applies universality by designing a neural network filter that can process multiple types of auxiliary information (prediction information, partitioning information, coding information) through a unified processing framework. This multi-functional approach allows the system to leverage diverse data sources without proportionally increasing complexity, as the same filter architecture handles different information types
Data Source
AI summary
Embodiments of the present disclosure provide a solution for video processing. A method for video processing is proposed. The method comprises: applying, for a conversion between a current video unit of a video and a bitstream of the video, a neural network filter to the current video unit at least based on auxiliary information associated with the current video unit, the auxiliary information including at least one of: prediction information of the current video unit, partitioning information of the current video unit, or coding information of a previously coded video unit; and performing the conversion based on the applying.