Template-Matched Affine Prediction for Complex Video Block Motion

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video coding technologies face challenges in improving coding efficiency and effectiveness, particularly in handling complex motion patterns beyond simple translations.

Innovation Solution

Implementing template matching (TM)-based refinement for affine motion candidates and control point motion vectors (CPMVs) to enhance affine inter prediction and advanced motion vector prediction (AMVP) in video processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional video coding techniques are used, then implementation is simpler, but coding efficiency is insufficient

Engineering Contradiction:

Improvecoding efficiencyVSAvoidprocessing complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The current video block is divided into multiple sub-blocks, and template matching is performed separately for each sub-block to determine sub-block motion vectors. This segmentation allows the system to handle complex motion patterns more effectively while maintaining manageable computational complexity through localized processing.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Template matching is applied selectively to sub-blocks rather than the entire video block. The system performs partial matching on regions where complex motion is detected or expected, avoiding full-template matching across all blocks, thus balancing coding efficiency improvement with computational complexity control.

Inventive Principle:

Principle #16Partial or excessive action

2Adaptability or versatility

If simple translation motion models are used, then processing is faster, but complex motion patterns cannot be captured

Engineering Contradiction:

Improvemotion pattern handlingVSAvoidmotion model complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system dynamically adapts the motion model complexity based on the content characteristics. For sub-blocks with simple motion, basic translation models are used; for sub-blocks with complex motion patterns, the system employs more sophisticated template matching and multi-hypothesis approaches, allowing the motion model complexity to vary dynamically across different regions and frames.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

Different motion modeling approaches are applied to different sub-blocks based on their local characteristics. Sub-blocks exhibiting complex motion patterns receive more sophisticated processing with template matching, while sub-blocks with simple motion use standard prediction methods, ensuring that computational resources are allocated where they are most needed.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS20250386006A1Method, apparatus, and medium for video processing

Publication Date: 2025.12.18 DOUYIN VISION CO LTD

AI summary

Embodiments of the present disclosure provide a solution for video processing. A method for video processing is proposed. In the method, for a conversion between a current video block of a video and a bitstream of the video, at least one of: an affine merge candidate of the current video block, an affine inter prediction of the current video block, or an affine advanced motion vector prediction (AMVP) of the current video block is determined, the current video block being coded with an affine mode. A template matching (TM)-based refinement is applied to the at least one of the affine merge candidate, the affine inter, or the affine AMVP. The conversion is performed based on the refinement.