Rice Parameter Derivation for Transform-Skip Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding techniques, such as those in HEVC and VVC, face inefficiencies in transform-skip coefficient coding, particularly in lossless mode where large coefficient residuals require large Rice parameter values for efficient representation, which existing schemes do not adequately address.
Innovation Solution
The proposed solution involves deriving a Rice parameter based on the sum of neighboring residual coefficients and using an offset to optimize Rice parameter look-up tables, allowing for efficient encoding and decoding of transform-skip coefficients in both lossless and lossy modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing Rice parameter derivation schemes are used for transform-skip coefficient coding, then the coding process is simple, but large coefficient residuals cannot be efficiently represented in lossless mode
Solution Approach 1:
The patent changes the parameters used for Rice parameter derivation by incorporating the sum of neighboring residual coefficients and applying an offset based on base level values. This allows the Rice parameter to adapt to large coefficient residuals in lossless mode while maintaining a systematic derivation process that does not significantly increase complexity.
2Productivity
If Rice parameter values are increased to handle large coefficient residuals, then representation efficiency improves, but the coding complexity and computational overhead increase
Solution Approach 1:
The patent performs preliminary computation of the sum of neighboring residual coefficients before deriving the Rice parameter. This pre-computation allows for more informed Rice parameter selection that handles large residuals efficiently, while the systematic approach using offsets and look-up tables keeps the overall coding complexity manageable.
Solution Approach 2:
The patent introduces an intermediate offset value that is added to the sum of neighboring coefficients before determining the final Rice parameter. This intermediary step allows for flexible adjustment of the Rice parameter based on base level values, improving compression efficiency without requiring complex direct calculations.
3Adaptability or versatility
If transform-skip coefficient coding uses fixed Rice parameter schemes, then the encoding process is fast, but adaptability to different residual patterns is poor
Solution Approach 1:
The patent makes the Rice parameter dynamic by deriving it from neighboring residual coefficients rather than using fixed values. The sum of neighboring coefficients and the offset-based adjustment allow the Rice parameter to adapt to different residual patterns in the video data, improving versatility while maintaining relatively fast encoding through systematic derivation.
Data Source
AI summary
Embodiments are directed to systems and methods of using a Rice code in video coding. In one embodiment includes a method of encoding or decoding video data, e.g., on a video encoder or decoder. In particular, embodiments are directed to determining a Rice parameter for encoding or decoding syntax elements related to coefficient levels.