Probability Estimator Update in Video Entropy Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies, such as HEVC, face limitations in achieving significant compression capabilities beyond existing standards, particularly in efficiently updating probability estimates during entropy decoding, which affects the accuracy and speed of binary arithmetic coding.
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, and performing bit-shifting operations to compute and update sub-intervals during entropy decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional probability updating methods are used in entropy decoding, then the decoding process is simpler, but the compression efficiency is insufficient
Solution Approach 1:
The patent applies parameter changes by modifying the probability estimate through conditional addition of 2N-d when the bin value equals the least probable state. This parameter adjustment optimizes the probability model adaptation speed (controlled by parameter d) and precision (controlled by bit width N), thereby improving compression efficiency without requiring complete redesign of the entropy decoding architecture.
2Measurement precision
If probability estimate precision is increased, then the accuracy of binary arithmetic coding improves, but the computational overhead increases
Solution Approach 1:
The patent implements partial action by selectively updating the probability estimate only when the bin value equals the least probable state, rather than updating after every bin. This conditional update approach (applying the 2N-d addition only when needed) maintains sufficient probability model accuracy while reducing unnecessary computational operations, thus balancing precision with computational power consumption.
3Speed
If the probability updating speed is increased, then the decoding throughput improves, but the accuracy of probability estimation may deteriorate
Solution Approach 1:
The patent applies dynamics by making the probability update adaptive through the parameter d, which controls the updating speed. The conditional update mechanism (updating only when bin value equals least probable state) creates a dynamic probability model that adapts to the actual data distribution, allowing faster updating when needed while maintaining accuracy by avoiding unnecessary updates that could introduce errors.
Data Source
AI summary
A non-transitory machine-readable medium of an electronic device storing computer-executable instructions for updating a probability estimator value during entropy decoding for a bitstream representing a set of video pictures is provided. The computer-executable instructions, when executed by a processor of the electronic device, update a probability estimator value by (i) performing an initial right bit-shifting operation on the probability estimator value to reduce a length, in bits, of the probability estimator value, (ii) multiplying the right bit-shifted probability estimator value by a range value representing an interval, (iii) performing another right bit-shifting operation on a result of the multiplication generated by multiplying the right bit-shifted probability estimator value by the range value, and (iv) adding a constant value to a result of the other right bit-shifting operation, wherein the probability estimator value is associated with a probability of a bin having a particular value.