Adaptive Merge Candidate List Generation for Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current inter-picture prediction coding methods in video compression, such as HEVC, face challenges in efficiently generating merge candidate lists for motion vector prediction, particularly in adapting between history-based and non-adjacent motion vector predictors based on block size, which affects encoding and decoding efficiency.
Innovation Solution
A method that determines the block size of a current block and selectively uses either history-based or non-adjacent motion vector predictors to generate a merge candidate list, allowing for adaptive switching between these predictors based on a threshold, thereby enhancing the encoding or decoding of video sequences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If history-based motion vector predictors are used for all block sizes, then encoding simplicity is maintained, but compression efficiency deteriorates for large blocks
Solution Approach 1:
The patent implements dynamic selection of motion vector predictor types based on block size. The system transitions from static use of history-based predictors to adaptive switching between history-based and non-adjacent predictors according to the current block's dimensions, optimizing compression efficiency while maintaining encoding simplicity through automated decision logic
Solution Approach 2:
The patent changes the parameter of predictor selection based on block size parameters. When block size exceeds a threshold, the system switches from history-based predictors to non-adjacent predictors, utilizing parameter-driven adaptation to resolve the contradiction between encoding simplicity and compression efficiency
2Loss of energy
If non-adjacent motion vector predictors are used for all block sizes, then compression efficiency improves, but device complexity increases
Solution Approach 1:
The patent applies local quality by using different predictor types for different block sizes. Instead of uniformly applying non-adjacent predictors to all blocks, the system selectively applies them only to large blocks where they provide benefit, while using simpler history-based predictors for smaller blocks, thereby reducing overall device complexity
Solution Approach 2:
The patent uses block size as a control parameter to determine predictor selection. This parameter-based approach automates the decision process, reducing the need for complex manual configuration or sophisticated selection algorithms, thus managing device complexity while maintaining compression efficiency gains
3Loss of energy
If adaptive switching between predictors is implemented, then compression efficiency improves, but encoding complexity increases
Solution Approach 1:
The patent implements adaptive switching through a simple parameter comparison: if block size exceeds a threshold, switch to non-adjacent predictors; otherwise, use history-based predictors. This parameter-driven approach achieves compression efficiency improvements while keeping encoding complexity manageable through automated, rule-based decision logic
Solution Approach 2:
The patent introduces dynamic adaptivity in predictor selection based on real-time block size characteristics. The system automatically adjusts its behavior according to the current block's dimensions, achieving compression efficiency improvements through adaptive switching while maintaining encoding complexity at acceptable levels through algorithmic automation
Data Source
AI summary
A method and apparatus for generating a merge candidate list for encoding or decoding of a video sequence includes determining a block size of a current block. History-based motion vector predictors and non-adjacent motion vector predictors are identified. The merge candidate list is generated using the history-based motion vector predictors and the non-adjacent motion vector predictors to permit the encoding or the decoding of the video sequence.