Residue Bin Context Modeling for Transform-Skip Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently encoding and decoding video data, particularly in managing context-coded bins and deblocking filters, which can lead to increased data redundancy and inefficiencies in compression.
Innovation Solution
The proposed method involves context modeling to determine context models for bins of syntax elements in transform skipped blocks, limiting the number of context-coded bins, and applying deblocking filters based on boundary strength and coding modes to improve encoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If context modeling is performed for all bins of syntax elements in transform skipped blocks, then decoding accuracy is improved, but computational complexity and processing time increase
Solution Approach 1:
The patent applies partial context modeling by selecting only a subset of bins for context-coded decoding rather than all bins. Specifically, it uses bypass-coded decoding for certain bins (e.g., significance flags, parity flags) while applying context-modelled decoding to other bins, thereby reducing computational complexity while maintaining adequate decoding accuracy for the most important syntax elements.
Solution Approach 2:
The patent segments the syntax elements in transform skipped blocks into different categories based on their importance and coding characteristics. It divides bins into those requiring context modeling and those suitable for bypass coding, allowing differential processing that balances accuracy and complexity requirements for different parts of the data structure.
2Reliability
If deblocking filters are applied with high boundary strength, then blocking artifacts are reduced, but processing time and computational load increase
Solution Approach 1:
The patent applies deblocking filters with differentiated boundary strength values based on local characteristics of block boundaries. Instead of uniformly applying strong filtering across all boundaries, it selectively applies higher boundary strength only where blocking artifacts are most problematic, while using lower or zero boundary strength in regions where filtering is less necessary, thereby reducing overall processing time while maintaining visual quality.
Solution Approach 2:
The patent dynamically adjusts the boundary strength parameter of deblocking filters based on local image characteristics, block types, and coding modes. By changing the boundary strength parameter adaptively rather than using a fixed high value, it achieves effective artifact reduction with reduced computational load compared to uniform strong filtering.
3Measurement precision
If transform skipped blocks use context-coded decoding for all bins, then reconstruction accuracy is improved, but bitrate increases
Solution Approach 1:
The patent implements partial context-coded decoding by applying context modeling only to specific bins within transform skipped blocks rather than all bins. It identifies which bins benefit most from context modeling (those with higher predictability) and applies context-coded decoding selectively, while using more compact bypass coding for other bins, thereby reducing bitrate while preserving reconstruction accuracy for critical syntax elements.
Data Source
AI summary
A method of video encoding includes performing context modeling to determine a context model for each of a number of bins of syntax elements corresponding to residues of a region of a transform skipped block in a current picture. The number of the bins of syntax elements being context coded does not exceed a maximum number of context coded bins set for the region. The method further includes encoding, according to Block Differential Pulse Code Modulation (BDPCM), the syntax elements based on the determined context models. When the maximum number of context coded bins is reached, remaining bins of syntax elements are encoded based on a bypass model.