Context-Adaptive CABAC Probability Updates for Stable Convergence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies using context adaptive binary arithmetic coding (CABAC) face challenges in maintaining probability update stability and convergence rate due to the use of a fixed probability update model, which fails to adapt to time-varying occurrence probabilities.
Innovation Solution
Implementing a method that includes context model determination, probability update, and probability interval determination for CABAC, with adaptive context model selection and multiple probability update methods, such as table-based, momentum-based, and boundary-based updates, to enhance compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If a fixed probability update model is used in CABAC, then the stability of probability update is improved, but the convergence rate deteriorates and cannot adapt to time-varying occurrence probabilities
Solution Approach 1:
The patent applies dynamics by transitioning from a fixed probability update model to a dynamic adaptive model. The context model determines whether to use table-based probability update (for stability) or operation-based probability update (for adaptability), allowing the system to dynamically adjust its behavior based on current coding conditions and symbol statistics.
Solution Approach 2:
The patent changes parameters by introducing multiple probability update methods with different characteristics. The system can switch between table-based update (with predefined probability tables) and operation-based update (with flexible probability calculations), effectively changing the update mechanism parameters to balance stability and adaptability.
2Device complexity
If only one probability update model is used, then the device complexity is reduced, but the compression efficiency deteriorates due to inability to reflect varying occurrence probabilities
Solution Approach 1:
The patent segments the probability update system into multiple distinct models (table-based and operation-based), each handling specific conditions. The context model divides the update process into different modes, allowing the system to select appropriate segments based on current requirements, thus maintaining manageable complexity while improving compression efficiency.
Solution Approach 2:
The patent implements multi-functionality by creating a unified context model that can handle multiple probability update scenarios. The context model serves multiple functions: determining which update method to use, managing probability intervals, and adapting to different symbol types, thereby improving compression efficiency without proportionally increasing complexity.
3Ease of manufacture
If a fixed probability table is used, then the implementation simplicity is improved, but the measurement precision of occurrence probability deteriorates
Solution Approach 1:
The patent introduces feedback mechanisms where the context model continuously monitors symbol occurrence statistics and adjusts the probability update accordingly. The system uses feedback from actual symbol occurrences to determine whether to switch between table-based and operation-based updates, improving measurement precision while maintaining implementation simplicity through automated adaptation.
Data Source
AI summary
There is provided an video encoding/decoding method and apparatus. The video decoding method comprises acquiring a bitstream including a predetermined context element, performing at least one of a context model determination, a probability update, and a probability interval determination on the predetermined syntax element, and arithmetically decoding the predetermined syntax element on the basis of a result of the performance.