Transform Coefficient Coding Simplification in Video Encoders
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of transform coefficient coding in video encoders and decoders, particularly in High Efficiency Video Coding (HEVC) and Versatile Video Coding (VVC), is high due to the need for efficient entropy coding and quantization, which limits processing speed and efficiency.
Innovation Solution
The proposed method simplifies the coding of transform coefficients by determining a Rice parameter based on neighboring coefficients, using a dependent quantization state, and employing Golomb-Rice coding to generate and parse coded bits, allowing early termination of subblock coding passes and bypass mode encoding for remaining coefficients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional transform coefficient coding methods are used in HEVC and VVC, then coding accuracy is maintained, but computational complexity increases and processing speed decreases
Solution Approach 1:
The patent extracts and processes only the most significant transform coefficients using a simplified coding method. By identifying and separately handling significant coefficients (those above a threshold) versus insignificant coefficients (those at or below the threshold), the method reduces computational complexity while maintaining coding accuracy for the most important data.
Solution Approach 2:
The patent changes the coding parameters dynamically based on the significance of coefficients. For significant coefficients, it uses a simplified coding approach with reduced precision, while for insignificant coefficients, it applies a different coding strategy. This parameter adaptation reduces overall computational complexity without significantly impacting reconstruction quality.
2Productivity
If simplified coding methods are used for transform coefficients, then processing speed increases, but coding accuracy may deteriorate
Solution Approach 1:
The patent applies different coding quality levels to different coefficients based on their significance. Significant coefficients (which have greater impact on reconstruction quality) are coded with sufficient precision to maintain accuracy, while insignificant coefficients are coded with reduced precision or omitted entirely. This local differentiation maintains overall coding accuracy while improving efficiency.
Solution Approach 2:
The patent applies full-precision coding only to the necessary subset of significant coefficients rather than all coefficients. By performing partial action (coding only what is necessary for acceptable quality) rather than excessive action (coding all coefficients at full precision), the method achieves better coding efficiency without significant loss in accuracy.
3Productivity
If conventional entropy coding is applied to all transform coefficients, then compression completeness is achieved, but bitstream parsing throughput decreases
Solution Approach 1:
The patent extracts significant coefficients from the full transform coefficient set and applies simplified entropy coding only to these extracted coefficients. Insignificant coefficients are handled through alternative methods (such as run-length encoding or omission with implicit zero reconstruction), reducing the number of coefficients requiring complex entropy decoding and thereby increasing parsing throughput while maintaining compression completeness.
4Manufacturing precision
If detailed coding is applied to all coefficients in a subblock, then reconstruction quality is maintained, but processing time increases
Solution Approach 1:
The patent changes the coding parameters (such as precision level, coding mode, and threshold values) based on the characteristics of each coefficient. By adapting parameters to the local content characteristics, the method maintains reconstruction quality where needed while reducing processing time for coefficients that contribute less to overall quality, achieving a better time-quality tradeoff.
Data Source
AI summary
Method and apparatus for encoding and decoding prediction residues in a video coding system also disclosed. At the decoder side, a Rice parameter for the target transform coefficient is determined based on a local sum of absolute levels of neighboring transform coefficients of the target transform coefficient. A dependent quantization state is determined and a zero-position variable is determined based on the dependent quantization state and the Rice parameter. One or more coded bits associated with a first syntax element for the target transform coefficient in a transform block are parsed and decoded using one or more codes including a Golomb-Rice code with the Rice parameter, where the first syntax element corresponds to a modified absolute level value of the target transform coefficient. An absolute level value of the target transform coefficient is derived according to the zero-position variable and the first syntax element.