Inter-view Motion Vector Prediction for Video Coding Efficiency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video signal coding technologies face inefficiencies in predicting inter-view motion vectors, leading to suboptimal coding efficiency and increased residual data transmission.

Innovation Solution

The method involves deriving an inter-view motion vector from inter-view motion vector candidates, prioritizing them, and using them for inter-view prediction, along with disparity vectors and temporal/spatial neighboring blocks, to enhance prediction accuracy and reduce residual data transmission.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If inter-view motion vector prediction is performed using conventional methods, then coding efficiency is maintained at current levels, but prediction accuracy is insufficient and residual data transmission increases

Engineering Contradiction:
Improveprediction accuracyVSAvoidresidual data transmission
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent performs preliminary classification of neighboring blocks to identify which blocks are coded with inter-view inter-prediction before deriving motion vectors. This preliminary action enables selective use of inter-view motion vectors only for blocks that can benefit from them, improving prediction accuracy while avoiding unnecessary processing for blocks where such vectors would not be applicable or beneficial.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent applies different motion vector derivation strategies to different local regions (neighboring blocks) based on their coding characteristics. Specifically, inter-view motion vectors are used for blocks coded with inter-view inter-prediction, while other blocks use conventional temporal or spatial motion vectors. This localized approach optimizes prediction accuracy for each block type without uniformly increasing complexity across all blocks.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If inter-view motion vectors are derived for all blocks, then prediction accuracy may improve, but processing complexity increases significantly

Engineering Contradiction:
Improveinter-view motion vector accuracyVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent performs inter-view motion vector derivation only partially - specifically only for blocks that are coded with inter-view inter-prediction mode. This partial application avoids the excessive complexity of deriving such vectors for all blocks, while still achieving accuracy improvements where they are most needed. The method selectively applies the complex operation only when beneficial.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent performs preliminary classification to identify blocks coded with inter-view inter-prediction before attempting to derive inter-view motion vectors. This preliminary step prevents unnecessary derivation attempts for blocks where such vectors would not be applicable, thereby reducing overall processing complexity while maintaining accuracy for the relevant blocks.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If multiple inter-view motion vector candidates are evaluated without priority ordering, then accurate vectors may be found, but computational complexity increases

Engineering Contradiction:
Improveinter-view motion vector accuracyVSAvoidprocessing efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent segments the set of inter-view motion vector candidates into different priority levels or categories. By organizing candidates hierarchically rather than evaluating them as a flat set, the method improves processing efficiency while maintaining the ability to find accurate vectors. High-priority candidates are evaluated first, reducing the average search time.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary organization of motion vector candidates into priority-ordered groups before the actual selection process. This preliminary arrangement enables more efficient searching and selection, improving productivity while preserving the ability to identify accurate motion vectors through systematic evaluation of candidates in order of likelihood.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2849441B1Method and apparatus for processing video signals
Publication Date: 2019.08.21 LG ELECTRONICS INC
  • EP2849441B1 patent drawingFigure 1
  • EP2849441B1 patent drawingFigure 2
  • EP2849441B1 patent drawingFigure 3

AI summary

Abstract: The method for processing video signals according to the present invention involves deriving an inter-view motion vector of the current texture block from one of inter-view motion vector candidates in consideration of a priority among the inter-view motion vector candidates of the current texture block, and performing inter-view inter-prediction for the current texture block using the derived inter-view motion vector. The present invention enables more accurate inter-view motion vector prediction from among various inter-view motion vector candidates, and thus a reduction in the amount of residual data being transmitted, thereby improving coding efficiency.