Affine Motion Compensation With Boundary-Only OBMC
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards like VVC and AVS3 face challenges in achieving optimal coding efficiency and reducing computational complexity, particularly with the use of overlapped block motion compensation (OBMC), which is inefficient for variable block sizes and introduces significant complexity in encoder and decoder designs.
Innovation Solution
An adaptive OBMC scheme is proposed, adjusting the OBMC application area and weights based on block size and coding modes, and limiting OBMC to boundaries between CUs and their neighbors, while disabling it for certain coding modes to reduce complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If OBMC is applied to all blocks in existing video coding standards, then motion compensation accuracy is improved, but computational complexity and encoder/decoder design complexity increase significantly
Solution Approach 1:
The patent applies OBMC selectively only to boundary blocks that are adjacent to other CUs, rather than uniformly to all blocks. This local application approach maintains motion compensation accuracy at block boundaries where it is most needed, while avoiding unnecessary complexity in regions where OBMC provides minimal benefit.
Solution Approach 2:
The patent segments the application of OBMC by identifying and separating boundary blocks from non-boundary blocks. Boundary blocks are subjected to OBMC processing, while non-boundary blocks use standard motion compensation, thereby dividing the complex global problem into manageable local cases.
2Productivity
If OBMC is applied uniformly to all block sizes, then coding efficiency is improved, but adaptability to variable block structures deteriorates
Solution Approach 1:
The patent introduces dynamic adaptation of OBMC to variable block sizes by determining whether each block is a boundary block based on its position and size. The OBMC process is adaptively applied only when appropriate for the specific block configuration, allowing the system to handle diverse block sizes and shapes efficiently.
Solution Approach 2:
The patent changes the application parameters of OBMC based on block characteristics. Instead of fixed uniform application, the method adjusts OBMC application based on block size, position, and boundary conditions, optimizing coding efficiency for each specific case.
Data Source
AI summary
A method for video decoding is provided. The method includes obtaining a plurality of sub-blocks within an affine coding unit (CU) of a video frame, where the plurality of sub-blocks include a plurality of boundary sub-blocks and a plurality of non-boundary sub-blocks. Further, the method may include applying both overlapped block motion compensation (OBMC) and affine secondary prediction (ASP) to at least one sub-block to generate a sub-block-based prediction for the affine CU or enabling the ASP for the plurality of non-boundary sub-block and disabling the OBMC for the plurality of non-boundary sub-blocks.


