Context Modeling Transform Coefficients Entropy Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently reducing redundancy in video signals while maintaining acceptable quality, particularly in managing transform coefficients and context models for entropy coding.
Innovation Solution
The method involves context modeling to determine context models for syntax elements related to transform coefficients, where a group of transform coefficients with different template magnitudes within a predetermined range share a same context model, or a transform coefficient uses the same context model for different template magnitudes within the range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If separate context models are used for transform coefficients with different template magnitudes, then entropy coding precision is improved, but device complexity and memory usage increase
Solution Approach 1:
The patent changes the parameter of template magnitude by grouping coefficients with different magnitudes into shared context models. Instead of creating separate context models for each template magnitude value, the invention modifies the grouping parameter to combine multiple magnitude ranges into single context models, thereby reducing the total number of context models while maintaining effective entropy coding through the grouping strategy
Solution Approach 2:
The patent merges multiple transform coefficients with different template magnitudes into shared context models. By combining coefficients that previously required separate context models into unified groups, the invention reduces the overall number of context models needed, directly addressing the contradiction between coding precision and device complexity
2Manufacturing precision
If more context models are maintained for different transform coefficient magnitudes, then coding accuracy is improved, but memory usage increases
Solution Approach 1:
The patent modifies the parameter of context model allocation by changing from one-to-one mapping (each magnitude gets its own model) to many-to-one mapping (multiple magnitudes share models). This parameter change in the allocation strategy reduces the quantity of context models stored in memory while preserving coding accuracy through intelligent grouping
Solution Approach 2:
The patent combines multiple transform coefficient groups with different template magnitudes into shared context models. This merging approach reduces the total number of context models that must be stored in memory, directly reducing memory usage while maintaining coding accuracy through the shared model structure
3Productivity
If separate context models are used for each transform coefficient template magnitude, then entropy coding efficiency is improved, but computational cost increases
Solution Approach 1:
The patent merges the processing of multiple transform coefficient groups into unified context model operations. By combining what would otherwise require separate processing steps into single shared context model operations, the invention reduces the total computational operations needed, thereby lowering computational cost while maintaining entropy coding efficiency
Solution Approach 2:
The patent creates universal context models that serve multiple transform coefficient groups with different template magnitudes. Each shared context model performs multiple functions by handling different coefficient groups, reducing the overall computational burden compared to having dedicated single-function context models for each magnitude
Data Source
AI summary
A method of video decoding in a decoder is provided. A first template magnitude of a first transform coefficient in a specific frequency region of the transform block is determined. The first template magnitude is a first single value representing magnitudes of a first local template of the first transform coefficient. A first context model is identified for coding the syntax element of the first transform coefficient, the first context model being shared with at least a second transform coefficient in the specific frequency region of the transform block, a second template magnitude of the second transform coefficient having a second single value that belongs to a first subinterval. A first bin of the syntax element of the first transform coefficient and a second bin of the syntax element of the second transform coefficient is determined, from the coded bits, based on the first context model.