Multi-table CABAC Lookup Table Segmentation for Video Entropy Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current entropy coding techniques for image and video coding, such as CABAC in HEVC, face challenges in reducing computational complexity and lookup table size while maintaining performance, particularly due to large lookup tables required for probability state indexing and range derivation.
Innovation Solution
The implementation of a multi-table based Context-Based Adaptive Binary Arithmetic Coder (CABAC) using multiple-parameter probability models with reduced lookup tables, where each model is updated using individual sets of probability states and derived using equations that reduce the size of the lookup tables, such as those in Tables 3 to 4h, and employing range derivation methods like those in Tables 5 to 9 to minimize table size.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional CABAC with single lookup table is used, then entropy coding performance is maintained, but lookup table size and computational complexity are large
Solution Approach 1:
The patent divides the single large lookup table into multiple smaller lookup tables (e.g., Table 0, Table 1, Table 2, Table 3) that store different segments of probability state data. Each table handles a specific range or type of probability states, reducing the memory access burden and overall table size while maintaining complete probability modeling coverage.
Solution Approach 2:
Different lookup tables store probability data optimized for specific local conditions or probability ranges. Each table is tailored to handle particular syntax elements or probability state ranges more efficiently, allowing the system to adapt locally rather than using a uniform approach throughout.
2Measurement precision
If multiple-parameter probability models are implemented, then coding accuracy is improved, but computational complexity increases
Solution Approach 1:
The multiple-parameter probability model is segmented into separate lookup tables, each handling specific parameter combinations. This allows the complex multi-parameter modeling to be broken down into manageable table lookups rather than requiring complex real-time calculations.
Solution Approach 2:
Probability values and range LPS values for multiple parameter combinations are pre-calculated and stored in the lookup tables during table generation. This preliminary computation eliminates the need for complex real-time calculations during actual entropy coding operations.
3Device complexity
If reduced lookup tables are used, then hardware implementation cost is reduced, but probability state coverage must be maintained
Solution Approach 1:
The probability state space is segmented across multiple lookup tables, with each table covering specific state ranges. The system maintains complete coverage by strategically dividing the 64 probability states across multiple smaller tables, ensuring all states are representable while reducing individual table sizes.
Solution Approach 2:
The set of multiple lookup tables collectively provides universal coverage for all probability states and syntax elements. Each table serves multiple purposes by handling different probability ranges and syntax element types, making the reduced table structure equally versatile as the original large table.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
A method and apparatus of entropy coding for a video encoder or decoder using multiple-table based Context-Based Adaptive Binary Arithmetic Coder (CABAC) are disclosed. In one embodiment, a current bin of a binary data of a current coding symbol is encoded or decoded according to a probability of a binary value of the current bin and the probability of the binary value is updated according to the binary value of the current bin for a next bin by using multiple-parameter probability models. Each multiple-parameter probability model is updated using at least one lookup table with the individual set of probability state as a table index to access contents of said at least one lookup table. In another embodiment, the range update is calculated for a range interval based on middle value of the range interval.