CABAC Range Derivation Using LPS Thresholds and Reduced Tables

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing entropy coding methods, such as CABAC in HEVC, suffer from large lookup tables that require significant memory and computational resources, leading to inefficiencies and potential drops in coding performance when down-sampled.

Innovation Solution

Implement a method for deriving LPS probability and range indices using mathematical operations or lookup tables, reducing the size of the rangeOne table by considering only probabilities up to 0.5 and optimizing parameter initialization for different contexts and quality parameters (QP) to minimize computational complexity while maintaining coding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If context adaptive binary arithmetic coding is used to achieve high compression ratios, then compression efficiency is improved, but decoding complexity and processing time increase due to range derivation operations

Engineering Contradiction:
Improvecompression efficiencyVSAvoiddecoding complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent pre-calculates and stores threshold values in lookup tables during initialization, so that during decoding the range derivation operation only requires simple table lookup and comparison operations instead of complex real-time calculations. This preliminary preparation of threshold values significantly reduces decoding complexity while maintaining high compression efficiency.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If range derivation operations are performed with high precision to maintain coding accuracy, then measurement precision is improved, but processing time increases

Engineering Contradiction:
Improvecoding accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent changes the parameter representation by using fixed-point arithmetic with predetermined precision levels instead of full floating-point precision. By adjusting the precision parameter to match the requirements of the compression task, the system achieves sufficient coding accuracy while dramatically reducing processing time through faster arithmetic operations.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent uses simplified threshold values that are pre-computed and stored, which can be quickly retrieved and applied during decoding. These pre-computed thresholds act as disposable lookup values that provide sufficient precision without the computational overhead of real-time high-precision calculations.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

3Productivity

If complex range derivation algorithms are used to achieve optimal compression, then compression ratio is improved, but computational resources and processing time increase

Engineering Contradiction:
Improvecompression ratioVSAvoidcomputational resources
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent replaces complex mechanical range derivation calculations with simpler threshold-based decision logic. By substituting heavy computational operations with lighter table lookup and comparison operations, the system maintains high compression ratios while significantly reducing computational resource consumption and energy usage.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP3652937B1Method and apparatus for range derivation in context adaptive binary arithmetic coding
Publication Date: 2026.05.13 HFI INNOVATION INC
  • EP3652937B1 patent drawingFigure 1(a)~2
  • EP3652937B1 patent drawingFigure 3
  • EP3652937B1 patent drawingFigure 4

AI summary

A method and apparatus of entropy coding of coding symbols using Context-Based Adaptive Binary Arithmetic Coder (CABAC) are disclosed. According to the present invention, a math formula is used for calculating the LPS range for deriving rangeOne value and rangeZero value. For example, the LPS range is derived using one or more mathematical operations comprising calculating a multiplication of a first value and a second value, in which the first value is related to (2 *the LPS probability index + 1) or the LPS probability index, while the second value is related to (2 *the range index + 1) or the range index. Besides, the LPS range can be derived by using a look-up-table including table contents corresponding to values of LPS range associated with a set of LPS probability indexes and a set of range indexes for encoding or decoding a binary value of the current bin.