Conditional Affine Candidate Generation for Complete Motion-Vector Lists
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding techniques fail to completely fill the candidate list for affine motion vector candidates, leading to issues during encoding and decoding.
Innovation Solution
A method to generate additional candidates, including zero motion vector candidates, derived candidates based on neighboring blocks, and temporal motion vector candidates, to ensure the candidate list is fully populated.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If only inherited and constructed affine candidates are generated, then the candidate list generation is simple, but the candidate list is not completely filled leading to encoding/decoding issues
Solution Approach 1:
The patent applies partial action by generating additional candidates (zero motion vector candidates, temporal motion vector candidates, and derived candidates) only when the candidate list is not full after the initial inherited and constructed candidates are added. This ensures the candidate list is completely filled without unnecessarily generating all possible candidates, thus improving reliability while controlling complexity.
2Reliability
If the candidate list is not completely filled, then the encoding and decoding process is faster, but the encoder or decoder may choose non-existent candidates causing errors
Solution Approach 1:
The patent applies preliminary action by proactively filling the candidate list with additional candidates (zero motion vector candidates, temporal motion vector candidates, and derived candidates) before the encoding and decoding processes begin. This ensures that when the encoder or decoder needs to select a candidate, the list is already completely filled with valid options, preventing errors while maintaining efficient processing.
3Reliability
If additional candidate generation steps are added, then the candidate list is completely filled, but the processing time increases
Solution Approach 1:
The patent generates additional candidates (zero motion vector candidates, temporal motion vector candidates, and derived candidates) only when necessary - specifically, only when the candidate list is not completely filled after the initial inherited and constructed candidates are added. This conditional approach ensures the candidate list is completely filled for reliability while avoiding unnecessary processing time when the list is already sufficient.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The techniques described herein relate to methods, apparatus, and computer readable media configured to encode or decode video data. A current block of video data is coded using affine prediction. A first set of candidates of a candidate list for the current block is generated, including determining one or more inherited candidates and deriving one or more constructed candidates. After generating the first set of candidates, it is determined whether the candidate list is full. Upon determining the candidate list is not full, the candidate list is filled by generating a second set of candidates of the candidate list, including one or more of generating one or more zero motion vector candidates, generating one or more additional derived candidates based on the plurality of associated neighboring blocks of the current block, and generating a temporal motion vector candidate based on a temporal collocated picture.