HMVP Candidate List Signaling for Video Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies, such as H.265/HEVC, face challenges in efficiently predicting motion vectors (MV) for inter-picture prediction, leading to suboptimal compression ratios and increased bandwidth requirements due to limitations in MV prediction mechanisms.
Innovation Solution
The proposed solution enhances MV prediction by utilizing a history-based MV prediction (HMVP) method that maintains a table of HMVP candidates and applies pruning to reduce redundancy, allowing for more effective derivation of motion vectors from previous coded blocks, thereby improving compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If history-based MV prediction (HMVP) method is applied, then compression ratio is improved, but device complexity increases due to maintaining and pruning candidate tables
Solution Approach 1:
The patent applies preliminary action by pre-establishing a history-based MV prediction (HMVP) candidate table that stores previously coded motion vectors before they are needed for prediction. This allows the system to leverage historical data for improving compression without adding complex real-time computation during the actual prediction process.
Solution Approach 2:
The patent extracts and utilizes only the essential information from historical motion vectors by maintaining a pruned candidate table. Instead of storing all historical data, the system extracts and keeps only the most relevant HMVP candidates that improve prediction accuracy, thereby reducing the complexity burden while maintaining compression benefits.
2Loss of energy
If MV prediction accuracy is improved through history-based methods, then bandwidth requirements are reduced, but processing time increases due to additional prediction calculations
Solution Approach 1:
The system performs preliminary action by pre-computing and storing HMVP candidates in a table structure that can be quickly accessed during decoding. This allows the system to achieve improved compression ratios and reduced bandwidth requirements while minimizing real-time processing time through efficient table lookup rather than complex calculations.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method of video decoding includes receiving a coded video bitstream including a current picture. The method includes determining a predetermined condition associated with signaling data included in the coded video bitstream. The method includes determining, based on the predetermined condition, a size of an index included in the signaling data for a candidate list of vector predictors based on a number of merge mode candidates and a number of intra block copy (IBC) candidates. The method includes constructing the candidate list with vector predictors. The method includes retrieving a vector predictor from the candidate list in accordance with the index that has a value that does not exceed the determined size of the index. The method further includes decoding the current block in accordance with the retrieved vector predictor.