Motion Vector Selection in Scalable Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In scalable video coding using the H.264/SVC standard, the selection of motion vectors is inefficient due to the need for extensive searches across large image regions, which consumes significant computational resources and memory, particularly when dealing with enhancement layers and their reference layers.
Innovation Solution
The system penalizes scores of blocks outside a defined region versus those within it during the search for the highest scoring match, optimizing the search process by caching a portion of the image and adjusting the search window based on global motion vectors, thereby reducing computational load and improving efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large search region is used to find the highest scoring match for motion vector selection, then matching accuracy is improved, but computational complexity and memory consumption increase significantly
Solution Approach 1:
The patent divides the search region into multiple sub-regions or blocks, and performs motion estimation separately in each block. This segmentation allows the system to maintain high matching accuracy by searching sufficient areas while reducing overall computational complexity by processing smaller, divided regions rather than one large region at once.
Solution Approach 2:
The patent applies different search strategies or scoring criteria to different local regions based on their characteristics. By adapting the search approach to local image content properties, the system achieves high matching accuracy in each local area while avoiding uniform exhaustive search across the entire large region, thus reducing computational burden.
2Measurement precision
If the search window is expanded to cover more image blocks, then motion vector accuracy is improved, but memory consumption increases
Solution Approach 1:
The patent extracts and caches only the essential image data required for motion estimation in the search window, rather than loading or processing the entire image. By taking out only the necessary portions of image data that contribute to motion vector calculation, the system maintains accurate motion estimation while significantly reducing memory consumption.
3Manufacturing precision
If extensive search is performed across all image regions, then coding quality is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary actions such as pre-calculating reference data, pre-organizing image blocks, or pre-establishing search patterns before the actual motion estimation process. This preliminary preparation enables faster processing during the main coding phase while maintaining high coding quality, effectively reducing processing time without sacrificing accuracy.
Solution Approach 2:
The patent applies partial search strategies where not all possible image regions are exhaustively searched, but rather a carefully selected subset that provides sufficient matching accuracy. By performing partial action on the most promising regions rather than excessive action across all regions, the system achieves good coding quality with reduced processing time.
Data Source
AI summary
From among blocks within a region of a first image, a search is performed for a highest scoring match to a block of a second image. Searching for the highest scoring match includes penalizing scores of blocks outside a portion of the region versus scores of blocks within the portion of the region. A motion vector is coded between the block of the second image and the highest scoring match.