Region-Based Adaptive Loop Filtering for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional adaptive loop filtering in video coding is inefficient due to high computational complexity and power consumption, particularly in schemes like QC_ALF that apply filters on a pixel-by-pixel basis, and lacks consideration for system complexity in rate-distortion optimization.
Innovation Solution
A region-based adaptive loop filtering approach that divides video data into M×N regions, allowing for merge flags and various scanning patterns to reduce computational complexity and filter switching, along with rate-distortion-complexity optimization for filter selection and region partitioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If pixel-by-pixel adaptive loop filtering is applied (QC_ALF), then filtering performance is improved, but computational complexity and power consumption increase
Solution Approach 1:
The patent divides the image into multiple regions (e.g., 4x4 blocks or larger) and applies a single filter to each region rather than switching filters pixel-by-pixel. This segmentation approach maintains filtering effectiveness while significantly reducing computational complexity and power consumption by eliminating frequent filter switching operations.
2Reliability
If multiple filter candidates are applied per block, then rate-distortion performance is improved, but system complexity increases
Solution Approach 1:
The patent introduces a lambda parameter in the rate-distortion-complexity optimization function that allows dynamic adjustment of the optimization criteria. By changing the lambda value, the system can adaptively balance between filtering performance and complexity, selecting the optimal filter without exhaustively evaluating all candidates, thus reducing system complexity while maintaining performance.
3Reliability
If rate-distortion optimization is applied without complexity consideration, then rate-distortion performance is optimized, but system resources are wasted
Solution Approach 1:
The patent modifies the optimization function to include a complexity term with a lambda parameter: min(J(D, R) + λ*C). This parameter change enables the system to consider computational complexity alongside rate-distortion metrics, preventing unnecessary resource consumption on complex filtering operations that would not yield proportional performance improvements.
Solution Approach 2:
Instead of exhaustively evaluating all possible filter candidates and region partitions, the patent applies partial optimization by using the rate-distortion-complexity function to select representative filter candidates. This partial action approach achieves sufficient performance improvement while avoiding the excessive computational resources that would be required for complete enumeration of all possibilities.
Data Source
AI summary
In a block-based motion compensated system, coding noises may arise due to lossy operations. Adaptive loop filter (ALF) is a technique used to improve the quality. A pixel adaptive (PA) ALF method was developed that uses the Sum-modified Laplacian Measure (SLM) to classify pixels and applies a respective ALF filter to each pixel according to the SLM. While the PA ALF achieves better performance over a conventional single filter (SF) based ALF, the PA ALF causes higher complexity and consumes more power due to the high complexity and per-pixel filter switching. Accordingly region based ALF scheme is disclosed which allows adaptive filter selection on a region by region basis and does not require per-pixel filter switching. In one embodiment according to the present invention, a picture is divided into M×N regions. The region based ALF also allows region merging for the M×N regions to improve performance. In another aspect of the present invention, the optimal ALF design also takes into consideration of the system complexity by using rate-distortion-complexity optimization (RDCO). The RDCO technique is also applied to sample adaptive offset (SAO) design.