Adaptive OBMC Line Selection for Lower-Complexity Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face inefficiencies in Overlapped Block Motion Compensation (OBMC) due to increased computation complexity and memory bandwidth requirements, particularly when combined with Bi-Directional Optical Flow (BIO), leading to visual artifacts and reduced coding efficiency.
Innovation Solution
Adaptive OBMC methods that dynamically adjust the number of lines and weightings based on block characteristics, MV differences, boundary matching costs, and luminance data to optimize OBMC processes, reducing complexity and bandwidth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If OBMC is applied with fixed number of lines and weightings, then processing simplicity is maintained, but coding efficiency is reduced and visual artifacts increase
Solution Approach 1:
The patent applies dynamics by making the OBMC parameters (number of lines and weightings) adaptive rather than fixed. The encoder dynamically selects different numbers of lines and weighting values based on block characteristics such as motion vector differences and boundary conditions, allowing the system to optimize coding efficiency for each specific block while managing computation complexity through conditional logic.
Solution Approach 2:
The patent implements local quality by applying different OBMC configurations to different blocks based on their local characteristics. Each block can have customized line counts and weighting schemes tailored to its motion properties and boundary conditions, ensuring that regions with high motion variation receive more aggressive OBMC treatment while stable regions use simpler processing.
2Reliability
If OBMC is applied with increased number of lines and adaptive weightings, then visual quality improves, but memory bandwidth requirements increase
Solution Approach 1:
The patent changes parameters (number of lines and weighting values) based on block characteristics to optimize visual quality. By adjusting these parameters adaptively rather than using fixed high values, the system achieves improved visual quality where needed while reducing memory bandwidth consumption in regions where aggressive OBMC is not necessary.
3Productivity
If OBMC is applied with adaptive parameters, then coding efficiency is enhanced, but processing complexity increases
Solution Approach 1:
The system dynamically adapts OBMC parameters based on block characteristics, achieving improved coding efficiency through context-aware processing. The adaptive mechanism uses simple conditional checks on motion vector differences and block properties to determine parameters, avoiding complex optimization algorithms while still achieving significant coding efficiency improvements over fixed-parameter approaches.
Data Source
AI summary
A method and apparatus for Overlapped Boundary Motion Compensation (OBMC). A current MV (Motion Vector) for the current block/subblock is determined and a neighbouring MV for a neighbouring block/subblock associated with the current block/subblock is determined. A number of lines to be applied for OBMC (Overlapped Block Motion Compensation), one or more weightings associated with the OMBC for the number of lines, or both are determined according to one or more boundary matching costs between the current block/subblock and the neighbouring block/subblock, MV difference between the current MV and the neighbouring MV, or predicted luminance data associated with the current block/subblock. The OBMC is applied to at least one of the current block/subblock and the neighbouring block/subblock according to OBMC parameters comprising the number of lines, said one or more weightings, or both.