Transform Unit Context Derivation for Video Entropy Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing context index derivation scheme for transform split flags in video coding systems, such as HEVC, is complex and can result in unwanted contexts, especially when the LCU size is not equal to 64x64, leading to inefficiencies in entropy coding.
Innovation Solution
A new context index derivation scheme based on the transform unit size is proposed, which reduces the number of contexts from four to three, simplifying the entropy coding of transform split flags and decoupling the effects of CU and TU sizes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the existing context index derivation scheme is used for transform split flags in HEVC, then the coding can handle various LCU sizes, but the complexity of entropy coding increases and unwanted contexts are generated
Solution Approach 1:
The patent extracts and removes the problematic interaction between LCU size and context index derivation. By decoupling the context index calculation from LCU size considerations and basing it solely on transform unit size, the unwanted contexts are eliminated while preserving the ability to handle various LCU sizes through the transform unit partitioning structure.
Solution Approach 2:
The patent segments the context modeling into distinct transform unit-specific contexts rather than having a unified context model that must accommodate all LCU sizes. This segmentation allows each transform unit size (32×32, 16×16, 8×8, 4×4) to have its own dedicated context index, eliminating cross-contamination between different LCU size scenarios.
2Adaptability or versatility
If the existing context index derivation scheme is used, then comprehensive context coverage is achieved, but the number of contexts increases to four
Solution Approach 1:
The patent removes the redundant fourth context that arises from the interaction between LCU size and transform unit size in the existing scheme. By extracting only the transform unit size as the determining factor for context index, the scheme achieves comprehensive coverage of all transform unit scenarios with only three contexts (for 32×32, 16×16, and 8×8 transform units).
Solution Approach 2:
The patent changes the parameter used for context index derivation from a composite of LCU size and transform unit size to solely transform unit size. This parameter change reduces the state space from four possible contexts to three, eliminating the redundancy while maintaining adequate modeling coverage for all transform operations.
3Reliability
If the context index derivation is based on both LCU size and transform unit size, then all scenarios are covered, but the coupling of LCU and TU size effects creates inefficiency
Solution Approach 1:
The patent segments the context modeling responsibility from the LCU level to the transform unit level. By making context index derivation dependent only on transform unit size, the scheme achieves reliable coverage of all coding scenarios through the transform unit partitioning hierarchy while eliminating the inefficiency caused by coupling LCU and TU size effects in the context calculation.
Data Source
AI summary
Systems, apparatus, articles, and methods are described including operations for size based transform unit context derivation. In an example encoder, first circuitry is to encode video input data into a bitstream according to a bitstream syntax, wherein the video input data includes one or more pictures, the one or more pictures are partitioned into one or more coding tree blocks, the one or more coding tree blocks are partitioned into slices including one or more coding tree blocks, the one or more coding tree blocks include one or more transform blocks according to a transform tree including a split_transform_flag indicative of the split of a given coding block into corresponding one or more transform blocks, the split_transform_flag is coded using CABAC, and a context index associated with the CABAC coding of the split_transform_flag is based on a value. Second circuitry of the encoder is to output the bitstream.