Decoder-Side Motion Vector Derivation Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding techniques, such as those in HEVC, face challenges in efficiently deriving and utilizing motion vectors, leading to suboptimal coding efficiency and increased bit-rate due to the lack of effective constraints on motion information.
Innovation Solution
The implementation of decoder-side motion vector derivation (DMVD) with constraints on motion vectors and their differences, such as symmetry and anti-symmetry, to filter and select valid motion information, enhancing coding efficiency and reducing bit-rate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If decoder-side motion vector derivation (DMVD) is implemented without constraints, then motion vector accuracy may be improved, but coding efficiency decreases and bit-rate increases
Solution Approach 1:
The patent applies constraints on motion vector parameters (symmetry, anti-symmetry, magnitude limits) to filter derived motion vectors. This changes the parameter space by imposing mathematical relationships between motion vectors, thereby improving coding efficiency while maintaining acceptable accuracy through selective validation of derived vectors.
2Loss of information
If decoder-side motion vector derivation (DMVD) is implemented without constraints, then more motion information can be derived, but bit-rate increases
Solution Approach 1:
The patent extracts and applies specific constraints (symmetry, anti-symmetry, magnitude) from the full space of possible motion vector relationships. By taking out only the essential constraints needed for valid motion information, the patent reduces bit-rate while maintaining motion information completeness through selective validation rather than transmitting all possible motion data.
3Productivity
If constraints are applied to derived motion vectors, then coding efficiency improves, but the complexity of motion vector validation increases
Solution Approach 1:
The patent applies different types of constraints (symmetry, anti-symmetry, magnitude limits) to different motion vector scenarios based on local characteristics. This allows the validation process to focus on relevant constraints for each specific case rather than applying all possible constraints universally, thereby improving coding efficiency while managing validation complexity through context-dependent constraint selection.
Data Source
AI summary
Techniques related to decoder-side motion vector derivation (DMVD) are described. For example, this disclosure describes techniques related to applying one or more constraints to motion information, such as a motion vector (MV) derived by DMVD, and/or a MV difference between an initial MV and an MV derived by DMVD. When the constraint is applied to the DMVD, in certain examples, only the derived motion information which meets the constraint is regarded as valid motion information. Conditions may be placed on the constraints.