Video Coding Motion Vector Candidate Reordering for Lower Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing video coding standards, such as HEVC and VVC, face challenges in efficiently ordering motion vector predictor candidates, leading to increased bitrate and complexity due to non-representative candidate selection and redundant computations.
Innovation Solution
A method of generating and reordering motion vector predictor candidates by adding first and second sets based on availability and computed costs, excluding duplicates and zero candidates, to optimize the list for coding efficiency and reduce complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If motion vector predictor candidates are reordered based on computed cost, then coding efficiency is improved, but computational complexity increases
Solution Approach 1:
The candidate list is divided into two segments: candidates from the first set (spatial, temporal, historical) that are reordered based on computed cost, and candidates from the second set (zero candidates) that are excluded from reordering and placed at the end. This segmentation allows selective optimization without processing all candidates uniformly, balancing coding efficiency gains with computational complexity management.
2Productivity
If more motion vector predictor candidates are added to the list, then coding efficiency is improved, but bitrate increases
Solution Approach 1:
Zero candidates are added in advance to fulfill the maximum candidate number requirement before reordering occurs. By pre-positioning these zero candidates and excluding them from reordering, the system ensures the list reaches the required maximum size for optimal coding efficiency while preventing these candidates from consuming unnecessary bitrate through reordering operations.
3Adaptability or versatility
If duplicate candidates are included in the list, then candidate diversity is maintained, but computational redundancy increases
Solution Approach 1:
Duplicate candidates are identified and extracted from the reordering process. The patent specifically excludes duplicate candidates from the reordering operation while maintaining them in the list, thereby preserving candidate diversity for robust motion prediction while eliminating the computational redundancy of reordering identical candidates multiple times.
Data Source
AI summary
Improvements to the derivation and ordering of motion vector predictor candidates for predicting motion in an image portion are disclosed. A list of motion vector predictor candidates for predicting motion in an image portion is generated, in which a first set of motion vector predictor candidates to said list and a second set of motion vector predictor candidates is added to said list if the number of first set of motion vector predictor candidates is lower than a maximum candidate number so that the total number of candidates equals said maximum candidate number The list of candidates is reordered and said second set of candidates are excluded from said reordering.