Symbol-Adaptive Multi-Symbol Arithmetic Coding for Video Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current entropy coding techniques based on non-adaptive or frame-adaptive statistics are inefficient, especially for error-resilient operations, as they require explicit signaling of probabilities for each sub-frame element, leading to increased overhead and inefficiency in video compression.
Innovation Solution
The implementation of symbol-adaptive multi-symbol arithmetic coding techniques that allow for direct coding of non-binary values and aggregation of binary values, reducing the need for explicit probability updates and enhancing error-resilience by using context-adaptive methods to adapt multi-value symbols and eliminate forward adaptation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If non-adaptive or frame-adaptive statistics are used for entropy coding, then implementation complexity is reduced, but coding efficiency deteriorates and error-resilience is compromised
Solution Approach 1:
The patent implements symbol-by-symbol adaptive arithmetic coding where the probability model dynamically updates after each symbol is coded. This dynamic adaptation allows the encoder to track local statistics changes within frames, improving coding efficiency without requiring complex frame-level probability signaling, thus resolving the contradiction between implementation complexity and coding efficiency.
Solution Approach 2:
The patent segments the video stream into independently coded blocks where each block uses its own adaptive probability model. This segmentation allows error-resilient operation by isolating errors to individual blocks while maintaining high coding efficiency through local adaptation, avoiding the need for complex frame-level probability management.
2Measurement precision
If explicit probability signaling is performed for each sub-frame element, then coding accuracy is improved, but overhead increases and error-resilience deteriorates
Solution Approach 1:
The patent implements self-service adaptation where the probability model automatically updates based on observed symbol frequencies in each block. This eliminates the need for explicit probability signaling overhead while maintaining high coding accuracy through implicit adaptation, directly resolving the contradiction between coding accuracy and overhead.
3Productivity
If frame-based adaptation with explicit probability signaling is used, then coding performance is improved, but error-resilience and adaptability to sub-frame variations deteriorate
Solution Approach 1:
The patent divides the video stream into independently coded blocks with separate adaptive probability models. This segmentation provides error-resilience by isolating errors to individual blocks while maintaining high coding performance through local statistics adaptation, eliminating the need for frame-level probability signaling that compromises error-resilience.
Solution Approach 2:
The patent implements dynamic symbol-by-symbol adaptation within each block, allowing the probability model to quickly respond to local statistics changes. This dynamic approach maintains high coding performance while providing error-resilience by limiting the impact of errors to individual blocks rather than entire frames.
4Productivity
If symbol-by-symbol adaptation is implemented, then coding efficiency is improved, but implementation complexity and computational load increase
Solution Approach 1:
The patent implements self-service adaptation where the probability model automatically updates based on observed symbol frequencies without requiring complex external control mechanisms. This simple update rule achieves symbol-by-symbol adaptation efficiency while keeping implementation complexity manageable through straightforward probability recalibration after each symbol.
Data Source
AI summary
An encoder obtains an uncompressed video bitstream that includes syntax elements that are to be coded using multi-symbol alphabets such that at least one syntax element of the syntax elements that are to be coded determines a magnitude token level that is at least partially represented by a head part. For each of the at least one syntax element, the encoder encodes the head part using multi-symbol alphabets to include a magnitude token level and end-of-block (EOB) information. The encoder transmits to a decoder an encoded bitstream that includes syntax elements coded with the multi-symbol alphabets.