Affine Motion Search Subblock Skipping for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for affine decoder side motion vector refinement (DMVR) in video coding add complexity and require extensive computation for every subblock, leading to increased processing power requirements without significant coding efficiency gains.
Innovation Solution
A video coder is configured to skip at least one subblock of a block for affine motion compensation, excluding it from motion compensation processes and bilateral matching cost calculations, using a subset of subblocks to determine a final offset for affine DMVR mode.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all subblocks are used in affine DMVR process, then coding efficiency is maintained, but computation complexity increases significantly
Solution Approach 1:
The current block is divided into multiple subblocks, and only a selected subset of these subblocks is used for affine DMVR processing. This segmentation allows the system to process only relevant portions of the block, reducing computation complexity while maintaining coding efficiency for the most important regions.
Solution Approach 2:
Different subblocks are treated differently based on their importance. The patent applies affine DMVR selectively to specific subblocks that contribute most to coding efficiency, while skipping less important subblocks. This local differentiation optimizes the balance between computation cost and coding performance.
2Measurement precision
If extensive computation is performed for every subblock, then motion compensation accuracy improves, but processing time increases
Solution Approach 1:
Instead of performing full affine DMVR computation on all subblocks, the patent applies the process partially to only those subblocks that provide the most significant contribution to motion compensation accuracy. This partial action reduces processing time while maintaining sufficient accuracy for the overall block.
Solution Approach 2:
The patent extracts and processes only the essential subblocks that are most critical for accurate motion compensation, excluding less important subblocks from the computationally intensive affine DMVR process. This extraction approach maintains accuracy for key regions while reducing overall processing time.
3Device complexity
If subset of subblocks is used for affine DMVR, then computation complexity reduces, but coding efficiency may deteriorate
Solution Approach 1:
The patent applies affine DMVR selectively to specific subblocks based on their importance to coding efficiency. By identifying and processing only the most critical subblocks with high computational accuracy while skipping less important ones, the system maintains overall coding efficiency despite using a subset of subblocks.
Solution Approach 2:
The block is segmented into multiple subblocks with different importance levels. The patent processes the most important segments with full affine DMVR accuracy while using simplified or skipped processing for less important segments, thereby maintaining overall coding efficiency while reducing computation complexity.
Data Source
AI summary
Encoding and decoding video data using an affine decoder side motion vector derivation (DMVR) mode includes receiving a block of video data to be decoded using the affine DMVR mode, and dividing the block into a plurality of subblocks. A video encoder and video decoder may determine a final offset for the affine DMVR mode using a first subset of the plurality of subblocks. The video encoder and decoder may code the block of video data using the final offset to generate a coded block of video data.