Affine Motion Vector Clipping for Reliable Video Block Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video coding systems face issues with motion vectors that fall outside a certain range, leading to unintended consequences, particularly in block-based hybrid video coding systems.

Innovation Solution

Implement motion vector clipping for affine motion mode in video blocks, using control point affine motion vectors derived from neighboring blocks, and store clipped vectors for motion vector prediction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If motion vectors are used without clipping, then motion prediction accuracy is improved, but system reliability deteriorates due to unintended consequences from out-of-range values

Engineering Contradiction:
Improvemotion prediction accuracyVSAvoidsystem reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent applies parameter changes by clipping motion vectors to a valid range based on bit depth. The clipping operation modifies the parameter values (motion vector components) to ensure they fall within acceptable boundaries, preventing out-of-range values while maintaining prediction accuracy for valid vectors.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If motion vectors are clipped based on bit depth, then system reliability is improved, but motion prediction precision deteriorates due to loss of vector information

Engineering Contradiction:
Improvesystem reliabilityVSAvoidmotion prediction precision
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent applies local quality by selectively clipping only those motion vectors that fall outside the valid range, while preserving the full precision of vectors within the acceptable range. This localized approach ensures that valid motion information is maintained without introducing unnecessary clipping artifacts.

Inventive Principle:
Principle #3Local quality

3Device complexity

If control point affine motion vectors are derived from neighboring blocks, then device complexity is reduced, but measurement precision deteriorates due to potential error propagation

Engineering Contradiction:
Improvedevice complexityVSAvoidcontrol point affine motion vector precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies feedback by clipping the derived control point affine motion vectors to ensure they fall within valid ranges before being stored and used for prediction. This feedback mechanism prevents error propagation by correcting out-of-range values that could otherwise compromise prediction precision in subsequent processing stages.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3815375B1Adaptive control point selection for affine motion model based video coding
Publication Date: 2026.04.22 INTERDIGITAL VC HOLDINGS INC
  • EP3815375B1 patent drawingFigure 1A
  • EP3815375B1 patent drawingFigure 1B
  • EP3815375B1 patent drawingFigure 1C

AI summary

Systems, methods, and instrumentalities are disclosed for motion vector clipping when affine motion mode is enabled for a video block. A video coding device may determine that an affine mode for a video block is enabled. The video coding device may determine a plurality of control point affine motion vectors associated with the video block. The video coding device may store the plurality of clipped control point affine motion vectors for motion vector prediction of a neighboring control point affine motion vector. The video coding device may derive a sub-block motion vector associated with a sub-block of the video block, clip the derived sub-block motion vector, and store it for spatial motion vector prediction or temporal motion vector prediction. For example, the video coding device may clip the derived sub-block motion vector based on a motion field range that may be based on a bit depth value.