Binary Arithmetic Coding Interval Updates for Video Bitstreams
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding techniques, such as those in ITU-T H.264 and H.265, face limitations in efficiently updating probability estimates during binary arithmetic coding, which can impact the compression efficiency and bitstream generation in video encoding and decoding processes.
Innovation Solution
The method involves determining whether a bin value is equal to the least probable state and updating the probability estimate by conditionally adding 2N−d to the probability estimate, where N is the width in bits and d is a parameter controlling the probability updating speed, to enhance the probability estimation process in binary arithmetic coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional probability update methods are used in binary arithmetic coding, then the coding process is simple to implement, but the compression efficiency is limited
Solution Approach 1:
The patent modifies the probability update mechanism by introducing a conditional update rule based on bin value comparison with least probable state. The probability estimate is updated by adding 2^(N-d) only when the bin value equals the least probable state, where N is the probability width and d controls update speed. This parameter-based conditional update improves compression efficiency while maintaining implementation feasibility
Solution Approach 2:
The patent implements dynamic probability estimation by making the update process adaptive to the actual bin values encountered during coding. The probability estimate dynamically adjusts based on whether the current bin matches the least probable state, allowing the coding process to adapt to the statistical properties of the data being encoded, thereby improving compression performance
2Measurement precision
If probability estimation accuracy is increased, then compression performance improves, but computational complexity increases
Solution Approach 1:
The patent uses parameter d to control the probability updating speed, allowing adjustment of the balance between estimation accuracy and computational load. By conditionally adding 2^(N-d) based on bin value comparison, the method achieves improved accuracy without requiring complex computation, as the update rule is simple and deterministic
Solution Approach 2:
The patent applies partial updates to the probability estimate by only modifying it when specific conditions are met (bin value equals least probable state). This selective updating approach improves accuracy where needed while avoiding unnecessary computations when conditions are not met, thereby reducing overall computational complexity
Data Source
AI summary
Some embodiments provide a method for subdividing an interval during entropy coding for a bitstream representing a set of video pictures. The method performs a bit-shifting operation on a probability estimator value. The method computes a sub-interval value by multiplying a range value representing the interval by the bit-shifted probability estimator value. The method uses the sub-interval value to update the interval.