Context Modeling for LFNST Signaling in Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video encoding and decoding technologies face inefficiencies in reducing the signaling overhead of Low-Frequency Non-separable Transformation (LFNST) indices/flags, which affects the coding efficiency of context-adaptive binary arithmetic coding (CABAC) in video compression standards.
Innovation Solution
The introduction of new context models that allow for selective context-based coding of LFNST indices/flags based on parameters such as color component, block size, and number of non-zero transform coefficients, enabling more efficient signaling and coding of video data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single context model is used for CABAC coding of LFNST indices/flags, then the device complexity is reduced, but the coding efficiency and bitrate optimization are insufficient
Solution Approach 1:
The patent applies local quality by creating separate context models for different color components (luma and chroma). Instead of using a single context model for all LFNST indices/flags, the invention divides the context modeling into component-specific models that capture the unique statistical characteristics of each color component, thereby improving coding efficiency without requiring excessive complexity
2Quantity of substance
If multiple context models are used for different color components, then the bitrate is reduced and coding efficiency is improved, but the device complexity increases
Solution Approach 1:
The patent segments the context modeling into distinct models for different color components (luma and chroma). This segmentation allows each context model to be optimized for its specific component's statistical properties, enabling better compression and lower bitrate while keeping the complexity manageable through structured organization of the context models
3Productivity
If LFNST indices/flags are coded without context adaptation, then the ease of operation is improved, but the signaling overhead is reduced and coding efficiency is improved
Solution Approach 1:
The patent implements dynamic context adaptation where the context model is updated based on the actual data characteristics. The system dynamically adjusts the context probabilities during encoding/decoding operations, allowing efficient signaling of LFNST indices/flags while adapting to the specific patterns in the video data, thus reducing signaling overhead without losing information
Data Source
AI summary
An example method includes determining a color component of a unit of video data; determining, based at least on the color component, a context for context-adaptive binary arithmetic coding (CABAC) a syntax element that specifies a value of a low-frequency non-separable transform (LFNST) index for the unit of video data; CABAC decoding, based on the determined context and via a syntax structure for the unit of video data, the syntax element that specifies the value of the LFNST index for the unit of video data; and inverse-transforming, based on a transform indicated by the value of the LFNST index, transform coefficients of the unit of video data.