Residual Quadtree Coding for Video Transform Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards, such as HEVC, share the same quadtree decomposition structure for the luma and chroma components, which may not be optimal for each color component, leading to suboptimal transform unit partitioning and requiring additional split flags for independent decomposition.
Innovation Solution
The proposed solution involves signaling independent quadtree decomposition structures for each color component, where the context for entropy encoding a split flag for one component is based on the value of a split flag from another component, allowing for efficient signaling of additional split flags in the bitstream.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the same quadtree decomposition structure is shared for luma and chroma components, then the device complexity is reduced, but the transform unit partitioning becomes suboptimal for each color component
Solution Approach 1:
The patent segments the quadtree decomposition structure by introducing independent split flags for luma and chroma components. This allows each color component to have its own decomposition structure, resolving the contradiction by enabling optimized partitioning for each component while maintaining separate control mechanisms.
Solution Approach 2:
The patent applies local quality by allowing different decomposition characteristics for different color components. Each component (luma, chroma) can have its own split flag and decomposition pattern, enabling locally optimized transform unit partitioning rather than a uniform global structure.
2Manufacturing precision
If independent quadtree decomposition structures are used for each color component, then the transform unit partitioning optimality is improved, but the device complexity increases
Solution Approach 1:
The patent merges the decomposition control by using a unified quadtree framework that can accommodate both luma and chroma components. The context adaptive binary arithmetic coding (CABAC) mechanism combines the encoding of split flags for different components, reducing the overall complexity increase that would result from completely independent structures.
Solution Approach 2:
The patent creates a universal quadtree decomposition framework that serves multiple color components simultaneously. The same basic quadtree structure and CABAC encoding mechanism are used for both luma and chroma, allowing the system to handle multiple components with a single versatile infrastructure rather than requiring entirely separate systems.
3Adaptability or versatility
If additional split flags are signaled for independent decomposition, then the transform unit partitioning flexibility is improved, but the bitstream length increases
Solution Approach 1:
The patent implements feedback through context adaptive binary arithmetic coding (CABAC) where the context for encoding one split flag is determined by the value of another split flag. This adaptive encoding uses feedback from previously decoded flags to optimize the current encoding, reducing the bitstream length despite the increased number of flags.
Solution Approach 2:
The patent changes the encoding parameter by using context-adaptive variable-length coding instead of fixed-length coding for split flags. The context selection based on other split flag values allows the encoder to adapt the code length dynamically, achieving better compression efficiency while maintaining the flexibility of independent decomposition control.
Data Source
AI summary
A video decoding device receives an array of transform coefficients for a chroma component of video data. The video decoding device receives entropy encoded data representing the value of a split flag associated with the chroma component. The value of the split flag indicates whether the array of transform coefficients is divided into smaller transform blocks. The video decoding device determines a context for the entropy encoded data representing the split flag. The context is based on the value of a split flag associated with another component of video data. The video decoding device entropy decodes the data representing the value of the split flag based on the determined context using context adaptive binary arithmetic coding (CABAC). The luma and chroma components have independent residual quadtree (RQT) structures.