Motion Vector Difference Coding With Unified Context Modeling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video coding technologies face inefficiencies in entropy coding of motion vector differences due to high context numbers, leading to ineffective probability adaptation and increased coding complexity, as well as inadequate grouping of bins with varying symbol statistics.

Innovation Solution

Reducing the cutoff value for truncated unary coding to two bin positions, using a single context for both horizontal and vertical components, and employing advanced motion vector prediction methods, such as multiple predictors and merging of motion vector differences, to enhance coding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a high number of contexts is provided for coding motion vector differences, then the coding precision may be improved, but the coding complexity increases and probability adaptation fails to perform effectively

Engineering Contradiction:
Improvecoding precisionVSAvoidcoding complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges the context modeling for horizontal and vertical motion vector difference components into a single unified context. Instead of maintaining separate contexts for each component and bin position, the invention uses one shared context that adapts probability based on the absolute value of the motion vector difference, thereby reducing coding complexity while preserving coding precision through effective probability adaptation.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The single context is designed to serve multiple functions: it models both horizontal and vertical motion vector difference components, adapts to varying bin positions through the cutoff mechanism, and performs probability estimation for different absolute values of motion vector differences. This universal context replaces what would otherwise require multiple specialized contexts.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Measurement precision

If a high number of contexts is provided for coding motion vector differences, then the coding precision may be improved, but the execution of the decoding process is hampered due to necessary inspection of neighboring bins

Engineering Contradiction:
Improvecoding precisionVSAvoidease of decoding
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

By combining the context usage for horizontal and vertical components into a single shared context, the patent eliminates the need for the decoder to inspect neighboring bins to determine which context to use. The single context approach simplifies the decoding process while maintaining the ability to accurately model the statistics of motion vector differences.

Inventive Principle:
Principle #5Merging (Combining)

3Device complexity

If the number of contexts is provided too low, then the coding complexity is reduced, but bins of highly varying actual symbol statistics are grouped together and the probability estimation fails to effectively encode the bins

Engineering Contradiction:
Improvecoding complexityVSAvoidcoding precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies local quality by differentiating the treatment of bins based on the absolute value of the motion vector difference. Bins with absolute values below the cutoff use one probability estimation approach, while bins at or above the cutoff use another. This localized differentiation within the single context allows the system to handle varying symbol statistics effectively without requiring multiple full contexts.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The invention changes the parameter used for probability adaptation from the component sign (which would require separate contexts) to the absolute value of the motion vector difference. This parameter change enables a single context to adapt its probability estimation based on the magnitude of the difference, effectively handling bins with highly varying statistics while maintaining low complexity.

Inventive Principle:
Principle #35Parameter changes

4Measurement precision

If advanced motion vector prediction methods with multiple predictors are used, then the prediction quality is improved, but the amount of data to be transmitted increases

Engineering Contradiction:
Improveprediction qualityVSAvoiddata transmission volume
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent implements a list of multiple motion vector predictors but does not transmit all of them. Instead, it transmits only the index indicating which predictor from the list should be used. This partial transmission approach provides the benefits of multiple predictors for improved prediction quality while minimizing the data transmission volume by sending only the compact index rather than full predictor data.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3471415B1Entropy coding of motion vector differences
Publication Date: 2021.08.04 GE VIDEO COMPRESSION LLC
  • EP3471415B1 patent drawingFigure 1
  • EP3471415B1 patent drawingFigure 2A~2C
  • EP3471415B1 patent drawingFigure 3

AI summary

A decoder for decoding a video from a data stream into which horizontal and vertical components of motion vector differences are coded using binarizations of the horizontal and vertical components is described, the binarizations equaling a truncated unary code of the horizontal and vertical components, respectively, within a first interval of the domain of the horizontal and vertical components below a cutoff value, and a combination of a prefix in form of the truncated unary code for the cutoff value and a suffix in form of a Exp-Golomb code of the horizontal and vertical components, respectively, within a second interval of the domain of the horizontal and vertical components inclusive and above the cutoff value, wherein the cutoff value is two and the Exp-Golomb code has order one. An entropy decoder is configured to, for the horizontal and vertical components of the motion vector differences, derive the truncated unary code from the data stream using context-adaptive binary entropy decoding with exactly one context per bin position of the truncated unary code, which is common for the horizontal and vertical components of the motion vector differences, and the Exp-Golomb code using a constant equi-probability bypass mode to obtain the binarizations of the motion vector differences. A desymbolizer is configured to debinarize the binarizations of the motion vector difference syntax elements to obtain integer values of the horizontal and vertical components of the motion vector differences; A reconstructor is configured to reconstruct the video based on the integer values of the horizontal and vertical components of the motion vector differences.