Affine Motion Vector Prediction for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards, such as HEVC, face challenges in efficiently handling complex motion patterns within video sequences, particularly in describing two-dimensional block rotations and deformations, which can lead to increased bitrates and reduced coding efficiency.
Innovation Solution
The development of affine motion vector prediction techniques, including 4-parameter and 6-parameter affine models, which use control point motion vectors to generate affine MVP candidate lists, allowing for more accurate motion compensation and prediction, thereby improving coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If affine motion models are used to describe complex motion patterns, then motion description accuracy is improved, but bitrate increases
Solution Approach 1:
The patent changes the parameter representation from traditional motion vectors to affine transformation parameters (control point motion vectors). This allows complex motions like rotation and deformation to be described using a compact set of parameters, improving motion description accuracy while controlling bitrate through efficient parameter encoding.
Solution Approach 2:
The patent segments the motion description into control points that define affine transformations. Instead of describing every pixel's motion, only the control points' motion vectors are encoded, which then define the motion of the entire block through affine transformation, reducing bitrate while maintaining accuracy.
2Measurement precision
If affine MVP candidate lists are constructed with multiple candidates, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent performs preliminary construction of affine MVP candidate lists by deriving candidates from neighboring blocks and temporal reference blocks before the actual prediction process. This preliminary action organizes potential predictors in advance, improving prediction accuracy while managing computational complexity through structured candidate generation.
Solution Approach 2:
The patent applies different candidate selection strategies to different regions and block types. By constructing candidate lists with specific attention to local motion characteristics and block positions, the system improves prediction accuracy for each local region while avoiding uniform complex processing across the entire video stream.
3Adaptability or versatility
If more control points are used in affine models, then motion modeling capability is improved, but encoding overhead increases
Solution Approach 1:
The patent implements dynamic selection of affine model types (4-parameter or 6-parameter models with different numbers of control points) based on motion characteristics. This allows the system to adapt the modeling capability to the actual motion complexity in each block, improving versatility while controlling encoding overhead by using simpler models when sufficient.
Data Source
AI summary
Method and apparatus for coding system using affine motion model are disclosed. According to one method, a neighbouring block set of the current block comprising multiple spatial neighbouring blocks and one or more collocated blocks is determined for the current block. One or more constructed affine MVP candidates are derived for an affine MVP candidate list based on CP (control-point) MVs (motion vectors) at multiple spatial neighbouring blocks and said one or more collocated blocks. One constructed affine MVP candidate without one temporal MV is checked and inserted into the affine MVP candidate list before any constructed affine MV with one temporal MV. The current block or motion information of the current block is then encoded or decoded based on the affine MVP candidate list.