Video Processing Motion Candidate List Construction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video coding techniques, such as MPEG-2, MPEG-4, ITU-T.263, ITU-T.264/AVC, HEVC, and VVC, suffer from low coding efficiency, which is undesirable for digital video applications.
Innovation Solution
The proposed method determines different maximum numbers of motion candidates for motion vector and block vector candidate lists, and includes various candidate types like spatial, temporal, history-based, and spatial-temporal motion vector prediction candidates to construct a candidate list for improved video processing during conversion between video blocks and bitstreams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional video coding techniques are used, then implementation is simple, but coding efficiency is low
Solution Approach 1:
The motion candidate list is segmented into different types (spatial candidates, temporal candidates, HMVP candidates, pairwise candidates, STMVP candidates) with different maximum numbers. This segmentation allows selective construction of candidate lists based on prediction mode, improving coding efficiency while managing complexity through structured organization.
Solution Approach 2:
The maximum number of motion candidates is made dynamic based on the prediction mode being used. For MV prediction mode, the maximum number is set to one value, while for BV prediction mode, it is set to a different value. This dynamic adjustment optimizes coding efficiency for different prediction scenarios without requiring a fixed complex structure.
2Productivity
If different maximum numbers of motion candidates are determined for MV and BV candidate lists, then coding efficiency is improved, but processing complexity increases
Solution Approach 1:
Different maximum numbers of motion candidates are assigned to different prediction modes (MV mode vs. BV mode) based on their specific requirements. This local optimization allows each prediction mode to have tailored candidate list parameters, improving overall coding effectiveness while keeping the complexity manageable through mode-specific configuration.
Solution Approach 2:
The maximum number of motion candidates parameter is changed based on the prediction mode. For MV prediction mode, one maximum number is used, while for BV prediction mode, a different maximum number is used. This parameter change strategy enables optimized performance for different prediction scenarios without requiring fundamentally different data structures.
3Productivity
If multiple candidate types are included in the motion candidate list, then coding performance is improved, but construction complexity increases
Solution Approach 1:
The motion candidate list is segmented into five distinct candidate types (spatial, temporal, HMVP, pairwise, and STMVP candidates), each with specific construction rules and maximum numbers. This segmentation allows systematic construction of comprehensive candidate lists that improve video coding performance while managing complexity through structured, type-specific processing.
Data Source
AI summary
Embodiments of the present disclosure provide a solution for video processing. A method for video processing is proposed. The method comprises: determining, during a conversion between a target video block of a video and a bitstream of the video, a target maximum number from a first maximum number of motion candidates of a candidate type associated with a motion vector (MV) candidate list and a second maximum number of motion candidates of the candidate type associated with a block vector (BV) candidate list; determining a candidate list including motion candidates of the candidate type for the target video block based on the target maximum number; and performing the conversion based on the candidate list.