Geometric Partition Motion Refinement for Lower Video Coding Overhead
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards like VVC and AVS3 face inefficiencies in the geometric partition mode (GPM) due to suboptimal motion vector accuracy and increased signaling overhead, particularly when applied to non-merge inter modes, which limits coding efficiency.
Innovation Solution
The proposed method introduces motion vector refinement (GPM-MVR) for GPM partitions, using predefined MVD magnitudes and directions, and extends GPM to explicit inter modes with separate motion signaling (GPM-EMS), optimizing motion accuracy while minimizing signaling overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If motion vector refinement is applied to GPM partitions, then motion vector accuracy is improved, but signaling overhead increases
Solution Approach 1:
The video block is divided into two geometric partitions (first and second GPM partitions) with different motion characteristics. Each partition can independently apply template matching refinement, allowing selective refinement where needed while avoiding overhead in uniform regions. The partitioning enables localized motion vector refinement without requiring refinement across the entire block.
Solution Approach 2:
Different refinement strategies are applied to different partitions based on their local motion characteristics. The first GPM partition may use one refinement approach while the second uses another, allowing optimization tailored to local requirements. This prevents unnecessary refinement overhead in regions where simple motion models suffice.
2Productivity
If GPM is extended to explicit inter modes, then coding efficiency is improved, but device complexity increases
Solution Approach 1:
The GPM framework is designed to work across multiple inter prediction modes (merge mode and explicit inter mode) using a unified geometric partitioning approach. The same partitioning logic and motion vector derivation methods are applied universally, reducing the need for mode-specific complex logic while maintaining coding efficiency gains.
Solution Approach 2:
Different parameters and signaling mechanisms are used for different inter modes. In merge mode, motion vectors are derived from candidate lists, while in explicit inter mode, motion vectors are directly signaled. This parameter adaptation allows GPM to achieve coding efficiency in both modes without requiring fundamentally different processing complexity.
Data Source
AI summary
Methods and devices are provided for decoding a video block in GPM. The method includes: partitioning the video block into a first geometric partition and a second geometric partition; receiving a first template matching (TM) enable flag for the first geometric partition a second TM enable flag for the second geometric partition, the first TM enable flag indicating whether a uni-directional motion of the first geometric partition is refined by TM, and the second TM enable flag indicating whether a uni-directional motion of the second partition is refined by the TM; receiving a first merge GPM index for the first geometric partition and a second merge GPM index for the second geometric partition; constructing a uni-directional motion victor (MV) candidate list of the GPM; and generating a first uni-directional MV for the first geometric partition and a second uni-directional MV for the second geometric partition.