Affine Motion Vector Prediction Without Clipping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face limitations in efficiently handling affine motion compensation, particularly due to clipping operations that restrict motion vector magnitudes and the need for different precision levels for motion vector differences, which can increase bandwidth and computational complexity.
Innovation Solution
The proposed solution involves generating a list of affine motion vector prediction candidates without clipping operations and using multiple precision levels for motion vector differences, allowing for more flexible and efficient motion compensation and prediction, especially by deriving motion information at different block sizes and using bit shift operations for rounding motion vector differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If clipping operation is used to limit motion vector magnitude, then motion vector range is controlled, but coding efficiency deteriorates and bandwidth increases
Solution Approach 1:
The patent changes the parameter of motion vector precision by introducing multiple precision levels (e.g., 1/16th pixel, 1/32nd pixel, 1/64th pixel) for different motion vector difference components. This allows high precision where needed while maintaining coding efficiency by not applying uniform high precision across all motion vectors, thus resolving the contradiction between precision and coding efficiency.
2Measurement precision
If multiple precision levels are used for motion vector differences, then motion compensation accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies different precision levels to different motion vector difference components based on their specific needs. For example, horizontal motion vector differences may use one precision level while vertical components use another, or different precision levels are applied to different prediction candidates. This local differentiation improves accuracy where required while reducing overall computational complexity compared to uniform high precision.
3Reliability
If clipping operation is applied to affine motion vectors, then motion vector magnitude is restricted, but motion compensation flexibility deteriorates
Solution Approach 1:
The patent introduces dynamic precision selection for affine motion vectors, where the precision level is adapted based on the specific motion characteristics and block properties. Instead of static clipping, the system dynamically adjusts precision requirements, allowing high flexibility for complex motions while maintaining control where appropriate, thus resolving the contradiction between control and flexibility.
4Ease of manufacture
If uniform precision is used for all motion vectors, then implementation simplicity is maintained, but coding efficiency deteriorates
Solution Approach 1:
The patent segments the motion vector processing into multiple precision levels, dividing the previously uniform processing into distinct precision tiers. This segmentation allows the system to apply higher precision only where necessary (improving coding efficiency) while maintaining simpler processing elsewhere, thus balancing implementation simplicity with coding efficiency.
Data Source
AI summary
Video coding and decoding in which the bitstream is generated using affine motion compensation mode are described. One example includes generating, for a reconstruction of a current video unit, a list of affine motion vector prediction candidates in which each candidate in the list is calculated without using a clipping operation, wherein the clipping operation limits a magnitude of the candidate motion vectors to a range; and performing the reconstruction of the current video unit using the list of affine motion vector prediction candidates.