Video Coder Context Model Index Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face inefficiencies in entropy coding due to the reliance on neighboring block availability for determining context models, which can hinder optimal coding performance.
Innovation Solution
Determine context model indices for video encoding/decoding based on size-based constraints of current blocks, such as area, width, or height, to improve entropy coding efficiency by making conditional probabilities closer to 0 or 1.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If context model selection relies on neighboring block availability, then coding compatibility with existing standards is maintained, but entropy coding efficiency is suboptimal
Solution Approach 1:
The patent changes the parameter basis for context model selection from neighboring block availability flags to current block size parameters (width and height). This parameter change enables more predictable and efficient context model selection that directly correlates with block characteristics, improving entropy coding efficiency while maintaining implementation simplicity.
2Productivity
If context model index is determined by neighboring block flags, then implementation is straightforward, but coding performance is limited
Solution Approach 1:
The patent applies local quality by selecting context model indices based on the specific size characteristics of the current block. Different block sizes (width and height values) map to different context model indices, allowing the entropy coding to adapt to local block properties. This improves context probability accuracy by tailoring the context model to the actual block being coded rather than using a uniform approach based on neighboring blocks.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
This application discloses a processing method applied to video encoding or video decoding. The method includes: obtaining size information of a current block; determining a context model index ctxInc of a flag of the current block based on the obtained size information of the current block and a size-based constraint; and performing subsequent processing based on the context model index ctxInc of the flag of the current block. The method can be used to improve coding efficiency.