Image Encoding Candidate List Generation Using Motion Information Buffer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image encoding and decoding methods face inefficiencies in deriving motion information for prediction, particularly in generating MERGE/AMVP candidate lists, which affects the speed and accuracy of video compression.
Innovation Solution
The method involves generating a candidate list for a current block using spatial, temporal, and reconstructed information-based candidates, with the order and number of motion information added to the list determined by the inter prediction mode, and utilizing a buffer to store and retrieve motion information for effective prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional motion information derivation methods are used for generating MERGE/AMVP candidate lists, then the encoding/decoding process is simpler, but prediction efficiency and accuracy deteriorate
Solution Approach 1:
The patent applies preliminary action by pre-storing motion information from previously decoded blocks in a buffer before it is needed for generating MERGE/AMVP candidate lists. This allows the encoder/decoder to efficiently retrieve and utilize motion information from spatially or temporally adjacent blocks without performing complex real-time searches, thereby improving prediction accuracy while maintaining encoding simplicity.
2Productivity
If more motion information candidates are added to the candidate list, then prediction efficiency improves, but the amount of data to be processed increases
Solution Approach 1:
The patent applies local quality by selectively adding motion information candidates to the MERGE/AMVP candidate list based on their relevance and quality. Instead of uniformly processing all possible motion information, the method prioritizes adding candidates from spatially or temporally adjacent blocks that are most likely to provide accurate prediction, thereby improving prediction efficiency without proportionally increasing data processing volume.
3Ease of manufacture
If motion information is derived from reconstructed pictures, then the method is simpler to implement, but the prediction accuracy for current blocks deteriorates
Solution Approach 1:
The patent introduces an intermediary buffer that stores motion information from previously decoded blocks. This buffer acts as a mediator between the reconstructed pictures and the current block prediction process, allowing the system to utilize motion information from both reconstructed pictures and current picture blocks without directly complex interactions, thereby maintaining implementation simplicity while improving motion information accuracy.
Data Source
AI summary
A method and an apparatus for image encoding/decoding, according to the present invention, may generate a candidate list of a current block and perform inter-prediction of the current block by using any one of a plurality of candidates belonging to the candidate list. Here, the plurality of candidates comprise at least one from among a spatial candidate, a temporal candidate, and a restoration information-based candidate, and the restoration information-based candidate may be added from a buffer which stores motion information decoded prior to the current block.