Rice Parameter Derivation for High Bit-Depth Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In video coding standards like Versatile Video Coding (VVC), determining optimal Rice parameters for transform coefficients is challenging, especially at high bit-depths, due to limitations in existing methods that reduce compression efficiency and accuracy, particularly when coefficients are near the borders of transform blocks.
Innovation Solution
The proposed method updates the coefficient statistic value based on the encoding procedure used for each transform coefficient, considering context-based and absolute value encoding procedures, to derive a more accurate Rice parameter, enhancing coding efficiency by using a hybrid procedure for regular residual coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single Rice parameter derivation method is used for all transform coefficients, then the device complexity is reduced, but the coding efficiency and accuracy deteriorate, especially for high bit-depth video and border coefficients
Solution Approach 1:
The patent segments the transform coefficients into two groups: border coefficients (within 3 positions of right or bottom border) and non-border coefficients. Different Rice parameter derivation methods are applied to each segment, allowing optimized processing for each group while maintaining overall coding efficiency.
Solution Approach 2:
The patent applies different Rice parameter derivation strategies based on the local characteristics of coefficients. Border coefficients use one derivation method (considering spatial neighbors), while non-border coefficients use another method, matching the derivation approach to the local statistical properties of each coefficient group.
2Device complexity
If existing Rice parameter methods are used for high bit-depth video, then the device complexity is maintained, but the measurement precision and coding accuracy deteriorate
Solution Approach 1:
The patent introduces dynamic adaptation in Rice parameter derivation by considering the bit-depth of the video signal. For high bit-depth video (10-bit, 12-bit, etc.), the derivation process is adjusted to account for the different statistical distribution of transform coefficients, improving precision without requiring a complete redesign of the coding device.
3Productivity
If context-based encoding procedure is used for all coefficients, then the coding efficiency is improved, but the device complexity increases due to multiple encoding procedures
Solution Approach 1:
The patent applies context-based encoding selectively rather than universally. By identifying border coefficients and applying a simplified Rice parameter derivation method to them, the patent achieves partial optimization that balances coding efficiency improvement with device complexity management.
Data Source
AI summary
A method of decoding video data comprises updating a coefficient statistic value based on one or more transform coefficients of a transform block (TB), wherein updating the coefficient statistic value comprises, for each respective transform coefficient of the one or more transform coefficients of the TB: performing a derivation process to determine a temporary value, wherein the derivation process is determined based at least in part on which encoding procedure of a plurality of encoding procedures is used to encode the respective transform coefficient, the plurality of encoding procedures including a context-based procedure for encoding the respective transform coefficient and encoding the respective transform coefficient as an absolute value; and setting the coefficient statistic value as an average of the coefficient statistic value and the temporary value; determining a history value based on the coefficient statistic value; determining a Rice parameter for a specific transform coefficient of the TB.