Motion Vector Predictor Adaptation for Omnidirectional Video

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video compression methods for omnidirectional content, particularly those using equi-rectangular mapping, face inefficiencies due to geometric distortions and inadequate handling of motion vectors, leading to suboptimal compression performance and increased data requirements for encoding and decoding.

Innovation Solution

The method involves computing a scale factor for motion vector predictors, performing motion vector rescaling and transformation based on this scale factor, and using the transformed predictor for motion compensation and encoding or decoding of omnidirectional video data, specifically tailored for advanced motion vector prediction techniques.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If conventional motion vector prediction is used for omnidirectional video, then encoding complexity is reduced, but compression efficiency deteriorates due to geometric distortions

Engineering Contradiction:
Improvecompression efficiencyVSAvoidencoding complexity
Core Design Contradiction:
Loss of energyVSDevice complexity

Solution Approach 1:

The patent applies parameter changes by computing a scale factor based on temporal indices and picture resolutions, then using this scale factor to rescale motion vector predictors. This transforms the motion vector parameters to compensate for geometric distortions in omnidirectional video, improving compression efficiency while maintaining manageable encoding complexity through a systematic parameter adjustment approach.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If motion vector rescaling and transformation are performed, then precision of motion vector prediction is improved, but computational complexity increases

Engineering Contradiction:
Improveprecision of motion vector predictionVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements preliminary action by computing the scale factor and determining the rescaling/transformation order before actual motion compensation occurs. This advance preparation of transformation parameters enables higher precision motion vector prediction while organizing computational steps efficiently, thereby managing computational complexity through structured pre-processing.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If advanced motion vector prediction techniques are applied, then compression performance is improved, but data requirements for encoding and decoding increase

Engineering Contradiction:
Improvecompression performanceVSAvoiddata requirements
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent applies local quality by selectively transforming and rescaling motion vector predictors based on the computed scale factor and determined operation order. This localized adaptation of motion vector processing to specific temporal and spatial contexts improves compression performance by addressing geometric distortions where they occur, while avoiding unnecessary processing elsewhere, thus optimizing the balance between compression performance and data requirements.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS11076166B2Method and apparatus for motion vector predictor adaptation for omnidirectional video
Publication Date: 2021.07.27 INTERDIGITAL VC HOLDINGS INC
  • US11076166B2 patent drawing
  • US11076166B2 patent drawing
  • US11076166B2 patent drawing

AI summary

A method and apparatus adapts motion vector prediction for suitability to omnidirectional video. One embodiment improves handling of temporal motion vector predictors or rescaled motion vector predictors. Another embodiment is suited to spatial motion vector predictors, and another to a combination of either temporal or spatial motion vector predictors. The method analyzes a scale factor derived from, at least one of, the time index of the predictor, the time index of the reference image's predictor, the time index of a reference image's current block, and the time index of the current block. If, for example, the scale factor is greater than one, motion vector transformation is performed before motion vector rescaling. If, however, the scale factor is less than or equal to one, the motion vector rescaling is performed before motion vector transformation.