Arithmetic Coding Contexts for Non-Separable Transform Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face inefficiencies in context modeling for non-separable transforms, particularly in parallel processing of syntax elements related to transform coefficient levels, which affects coding efficiency.
Innovation Solution
A context model is derived for syntax elements associated with transform coefficient levels based on neighboring coefficient groups positioned on different scanning lines, using a reverse diagonal scan order to improve coding efficiency without disrupting parallel processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a context model is determined for syntax elements using traditional scanning methods, then coding efficiency can be improved through better probability modeling, but parallel processing capability is disrupted
Solution Approach 1:
The patent applies reverse diagonal scan order instead of traditional diagonal scan order to determine context models for syntax elements. This inversion of the scanning sequence allows the context modeling to follow a pattern that is compatible with parallel processing architectures, resolving the contradiction between improved coding efficiency through context modeling and maintenance of parallel processing capability.
Solution Approach 2:
The patent introduces a new scanning dimension by using reverse diagonal scan order across multiple scanning lines, rather than processing coefficients in the traditional sequential manner. This dimensional change in the scanning approach enables both effective context modeling and preservation of parallel processing capabilities by reorganizing the processing sequence without reducing the degree of parallelism.
2Measurement precision
If transform coefficient levels are coded using existing context models, then decoding accuracy is maintained, but coding efficiency is reduced due to suboptimal probability modeling
Solution Approach 1:
The patent changes the scanning order parameter from traditional diagonal scan to reverse diagonal scan when determining context models for transform coefficient levels. This parameter change in the scanning sequence improves the accuracy of probability modeling by better capturing the statistical dependencies between neighboring coefficients, thereby improving both decoding accuracy and coding efficiency simultaneously.
Solution Approach 2:
The patent uses transform coefficient levels from neighboring coefficient groups (determined through reverse diagonal scanning) as feedback to improve the context model for the current coefficient group. This feedback mechanism allows the decoder to use previously decoded neighboring values to inform the probability modeling of current coefficients, improving both accuracy and efficiency.
Data Source
AI summary
A video bitstream including a current transform block (TB) in a current picture is received. A context model is determined for a syntax element associated with a transform coefficient level of a first coefficient group (CG) in the current TB based on transform coefficient levels of at least one first neighboring CG of the first CG. The first CG is positioned on a first scanning line. The at least one first neighboring CG is positioned on a second scanning line that is scanned before the first scanning line. The context model is a probability model for a non-separable transform. The first CG is reconstructed based on the transform coefficient level that is determined according to the determined the context model.