Motion Vector Confidence Patch Priority List for Scalable Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motion estimation methods are resource-intensive and costly, making them impractical for consumer-level devices, while recent advancements have made them cost-effective, but still require efficient processing and memory bandwidth optimization for scalable video coding.
Innovation Solution
A system and method using motion vector confidence to determine a fine motion estimation patch priority list, where coarse motion vectors and their confidences are used to generate a patch list, sorting video blocks by confidence to optimize search areas for accurate fine motion vector determination, reducing processing and memory bandwidth requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional motion estimation methods are used, then motion estimation can be performed, but processing cost and resource consumption are excessively high
Solution Approach 1:
The motion estimation process is divided into two stages: coarse motion estimation at lower resolution and fine motion estimation at full resolution. This segmentation allows the computationally intensive fine motion estimation to be performed only on selected blocks, reducing overall processing cost while maintaining accuracy where needed.
Solution Approach 2:
Different levels of motion estimation accuracy are applied to different video blocks based on their importance and motion characteristics. High-motion or important blocks receive fine motion estimation, while other blocks use coarse motion estimation, optimizing the balance between quality and processing cost.
2Measurement precision
If fine motion estimation is performed on all video blocks, then motion estimation accuracy is maximized, but processing time and memory bandwidth requirements increase significantly
Solution Approach 1:
Coarse motion estimation is performed as a preliminary step before fine motion estimation. This preliminary action provides initial motion vectors that guide the subsequent fine motion estimation process, allowing the system to focus computational resources on blocks that need high precision while avoiding unnecessary processing on blocks where coarse estimation suffices.
Solution Approach 2:
Fine motion estimation is applied partially only to selected video blocks rather than all blocks. The selection is based on confidence metrics from coarse motion estimation and block importance, performing excessive action (fine estimation) only where necessary to achieve the desired overall accuracy.
3Adaptability or versatility
If motion estimation is performed in consumer-level devices, then video compression capability is improved, but processing resources are limited
Solution Approach 1:
The system dynamically changes processing parameters including resolution levels, search range sizes, and estimation precision based on available resources and block characteristics. This allows consumer-level devices to adapt motion estimation complexity to match their limited processing resources while still providing video compression capability.
Data Source
AI summary
Methods and systems for using motion vector confidence to determine a FME patch priority list for a scalable coder are disclosed, and may include a fine motion estimator receiving a plurality of coarse motion vectors and corresponding confidences. A patch list may be generated based on the corresponding confidences of the coarse motion vectors. The patch list may then be used to determine a search area. Each video block in a present picture may be matched to the video blocks in the search area to find the best match. A fine motion vector may be determined for each video block in the present picture with respect to a video block in the search area.