Geometric Video Block Partitioning for Accurate Motion Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards face challenges in efficiently utilizing geometry partitioning for improved motion vector prediction, leading to suboptimal compression efficiency and increased bandwidth demand.
Innovation Solution
Implementing geometry partitioning modes in video coding that allow for non-rectangular and non-square prediction partitions, enabling advanced motion vector prediction techniques such as bi-prediction with Coding Unit based Weighting (BCW) and multi-hypothesis techniques, along with context-based coding and adaptive motion vector resolution, to enhance inter prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If geometry partitioning modes are implemented for video coding, then motion vector prediction accuracy is improved, but device complexity increases
Solution Approach 1:
The video block is divided into multiple prediction partitions using geometry partitioning modes (triangular, polygon, or rectangular partitions). Each partition can have its own motion vector, allowing for more precise motion representation in different regions of the block while maintaining manageable complexity through structured partitioning patterns.
Solution Approach 2:
Different motion vectors are assigned to different prediction partitions within the same video block. This allows the coding system to adapt to local motion characteristics in different regions, improving prediction accuracy for blocks with complex or non-uniform motion patterns without requiring completely different coding approaches for each block.
2Productivity
If advanced motion vector prediction techniques are used, then compression efficiency is improved, but bandwidth requirements increase
Solution Approach 1:
Multiple motion vector prediction techniques (merge mode, advanced merge mode, and geometry-based partitioning) are combined into a unified coding framework. The encoder can select from multiple candidate motion vectors and partitioning patterns, choosing the most efficient combination for each block to maximize compression while controlling bit rate.
Solution Approach 2:
The coding system dynamically selects from multiple geometry partitioning modes (triangular, polygon, rectangular) and motion vector prediction strategies based on the characteristics of each video block. This adaptive approach allows the system to achieve high compression efficiency for complex blocks while using simpler modes for easier blocks, optimizing the overall bandwidth usage.
Data Source
AI summary
A method for processing a video includes performing a determination, by a processor, that a first video block is partitioned to include a first prediction portion that is non-rectangular and non-square; adding a first motion vector (MV) prediction candidate associated with the first prediction portion to a motion candidate list associated with the first video block, wherein the first MV prediction candidate is derived from a sub-block MV prediction candidate; and performing further processing of the first video block using the motion candidate list.