Video Inter-Prediction Merge List Redundancy Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video signal processing technologies face challenges in efficiently handling next-generation video contents with high spatial resolution, high frame rate, and high dimensionality, requiring significant increases in memory storage, memory access rate, and processing power.
Innovation Solution
A method and device for processing video signals using inter prediction, specifically constructing a merge list based on neighboring blocks, adding history-based merge candidates, and generating prediction blocks to reconstruct current blocks, while limiting redundancy checks to enhance efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If history-based merge candidates are added to the merge list without redundancy checks, then the prediction accuracy is improved, but the processing complexity increases
Solution Approach 1:
The patent changes the parameter of redundancy checking by introducing a threshold-based mechanism. Instead of performing exhaustive redundancy checks on all history-based merge candidates, the system checks redundancy only when the motion vector difference exceeds a predefined threshold. This selective parameter change reduces processing complexity while maintaining prediction accuracy for significant motion changes.
Solution Approach 2:
The patent applies partial action by performing redundancy checks only on a subset of history-based merge candidates rather than all candidates. Specifically, redundancy checking is performed only when motion vector differences exceed a threshold, leaving other candidates to be processed without extensive checks. This partial approach balances prediction accuracy with processing efficiency.
2Reliability
If comprehensive redundancy checks are performed for adding history-based merge candidates, then the merge list quality is improved, but the processing time increases
Solution Approach 1:
The patent modifies the time parameter by introducing a threshold-based redundancy check mechanism. Instead of performing time-consuming comprehensive checks on all history-based merge candidates, the system only performs checks when motion vector differences exceed a predefined threshold. This parameter change significantly reduces processing time while maintaining merge list quality for candidates with significant motion differences.
Solution Approach 2:
The patent applies skipping by bypassing redundancy checks for history-based merge candidates whose motion vector differences are within the threshold. This allows the system to rush through the processing of these candidates without extensive verification, significantly reducing processing time while still ensuring quality for candidates with larger motion differences that require verification.
3Adaptability or versatility
If all history-based merge candidates are added to the merge list, then the prediction coverage is improved, but the memory usage increases
Solution Approach 1:
The patent changes the quantity parameter by introducing a threshold-based filtering mechanism. Instead of adding all history-based merge candidates to the merge list, the system only adds candidates whose motion vector differences exceed a predefined threshold. This parameter change reduces memory usage by excluding redundant candidates with small motion differences while maintaining prediction coverage for significant motion changes.
Solution Approach 2:
The patent applies extraction by removing redundant history-based merge candidates from the merge list based on redundancy checks. Candidates with motion vector differences within the threshold are extracted (removed) from the candidate set, reducing memory usage. This extraction process maintains prediction coverage by retaining only the most significant motion candidates.
Data Source
AI summary
The present invention provides a method of processing a video signal based on inter prediction, the method comprises constructing a merge list based on a neighboring block of a current block; adding a history-based merge candidate representing motion information of a previously coded block of the current block to the merge list; obtaining a merge index indicating a merge candidate used for inter prediction of the current block within the merge list; generating a prediction block of the current block based on motion information of the merge candidate indicated by the merge index; and adding a residual block of the current block to the prediction block to reconstruct the current block, wherein when the history-based merge candidate has motion information that is not redundant with a pre-defined merge candidate among merge candidates included in the merge list, the history-based merge candidate is added to the merge list.