Adaptive Rice Parameter Derivation for Transform Coefficient Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-resolution, high-quality images and videos, particularly in virtual reality and augmented reality, requires more efficient compression techniques to reduce transmission and storage costs, improve residual coding efficiency, and derive more accurate rice parameters for transform coefficients.
Innovation Solution
A method and apparatus for deriving a rice parameter based on neighboring transform coefficients to enhance video/image coding efficiency, improve residual coding accuracy, and reduce codeword length after binarization in level coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional compression techniques are used for high-resolution content, then image/video quality is maintained, but transmission and storage costs increase significantly
Solution Approach 1:
The patent applies parameter changes by deriving Rice parameters dynamically based on neighboring transform coefficient statistics (mean, standard deviation) rather than using fixed or simply-derived parameters. This adaptive parameter adjustment enables more efficient entropy coding of transform coefficients, reducing the bitrate required to encode high-resolution content while maintaining reconstruction quality
Solution Approach 2:
The invention introduces dynamics by making the Rice parameter derivation process adaptive to local content characteristics. The Rice parameter is updated based on the statistical properties of neighboring coefficients, allowing the coding system to dynamically adjust to varying content complexity and directional patterns, thereby improving compression efficiency across different image regions
2Productivity
If existing Rice parameter derivation methods are used, then coding process is simple, but residual coding efficiency is insufficient
Solution Approach 1:
The patent implements self-service by having the Rice parameter derivation process automatically adapt to local content characteristics using statistical analysis of neighboring coefficients. The system derives its own optimal parameters based on the data at hand (coefficient statistics) without requiring external configuration or manual tuning, thereby improving residual coding efficiency while keeping the process self-contained
Solution Approach 2:
The invention applies preliminary action by computing statistical parameters (mean and standard deviation) of neighboring transform coefficients before deriving the Rice parameter. This preliminary statistical analysis enables the subsequent Rice parameter to be more accurately tailored to the specific content characteristics, improving coding efficiency before the actual entropy coding takes place
3Loss of information
If fixed Rice parameters are used for all coefficients, then encoding is fast, but codeword length increases
Solution Approach 1:
The patent applies local quality by deriving different Rice parameters for different coefficient positions based on their local statistical characteristics. Instead of using a uniform parameter across the entire block, the system analyzes neighboring coefficients locally and derives position-specific Rice parameters, resulting in shorter codewords for regions with specific statistical patterns while maintaining overall encoding speed
Data Source
AI summary
A video decoding method performed by a decoding apparatus comprises the steps of: acquiring, from a bitstream, residual information for a current transform block, wherein the residual information includes at least one of information indicating a level value of a target transform coefficient within the current transform block or information indicating a residual level value; based on a neighboring transform coefficient of the target transform coefficient, deriving a rice parameter for at least one of the information indicating the level value of the target transform coefficient or the information indicating the residual level value; based on the rice parameter, deriving a bin string for at least one of the information indicating the level value of the target transform coefficient or the information indicating the residual level value; and based on the bin string, deriving the level value of the target transform coefficient.