Affine Motion Vector Prediction for Non-Affine Video Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video coding techniques suffer from low coding efficiency, which is undesirable for modern digital video applications.
Innovation Solution
A method for video processing that derives a motion vector predictor from a parameter table storing affine parameters from previously coded blocks and performs conversions based on this predictor, improving coding efficiency and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional video coding techniques are used, then the video can be encoded/decoded, but the coding efficiency is low
Solution Approach 1:
The patent changes the parameters used for motion vector prediction by introducing affine motion models with multiple control points (CP0, CP1, CP2) and their corresponding motion vectors (MV0, MV1, MV2). This transforms the prediction from simple translational motion to affine motion that can model complex motion patterns, thereby improving coding efficiency and performance
Solution Approach 2:
The patent segments the motion prediction into multiple control points, where each control point has its own motion vector. This segmentation allows different regions of the block to have different motion characteristics, enabling more accurate motion modeling and improving the coding efficiency
2Productivity
If motion vector predictors are derived using affine models from previously coded blocks, then coding efficiency improves, but the complexity of the processing increases
Solution Approach 1:
The patent performs preliminary action by storing affine parameters from previously coded blocks in a parameter table before they are needed for prediction. This pre-computation and storage of motion vector predictors reduces the computational complexity during the actual prediction process, as the system only needs to retrieve and apply pre-derived parameters rather than computing them in real-time
Solution Approach 2:
The patent uses copying by deriving motion vector predictors for the current block by copying and adapting affine parameters from previously coded blocks. The system copies the affine model structure and applies it to predict motion vectors for blocks with similar characteristics, reducing processing complexity while maintaining high coding efficiency
Data Source
AI summary
Embodiments of the disclosure provide a solution for video processing. A method for video processing is proposed. The method includes: deriving, during a conversion between a target block of a video and a bitstream of the target block, a motion vector predictor for the target block from a parameter table that stores a set of affine parameters from at least one previously coded block, the target block being a non-affine coded block; and performing the conversion based on the motion vector predictor.