In-Loop Filtering for Video Encoding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding/decoding techniques struggle to efficiently minimize distortion between original and reconstructed pictures, particularly when using deblocking filtering and sample adaptive offset methods.
Innovation Solution
The proposed method employs in-loop filtering using subsampling-based block classification and multiple filter shapes to reduce computational complexity and memory requirements, thereby improving encoding/decoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If deblocking filtering is applied to reduce blocking artifacts around block boundaries, then blocking artifacts are reduced, but distortion between original picture and reconstructed picture increases
Solution Approach 1:
The patent divides the filtering process into multiple stages: first applying deblocking filtering to remove blocking artifacts, then applying sample adaptive offset filtering to correct residual distortion. This segmentation allows each filter to address specific issues without interfering with the other, resolving the contradiction between artifact removal and distortion minimization.
Solution Approach 2:
The patent applies different filtering characteristics to different regions of the image. The deblocking filter is applied strongly at block boundaries where blocking artifacts are most prominent, while the sample adaptive offset filter is applied selectively based on local gradient information and edge detection. This local differentiation allows optimal artifact reduction without excessive distortion in any single region.
2Manufacturing precision
If sample adaptive offset is applied to reduce ringing artifacts, then distortion is reduced to some degree, but computational complexity and memory access bandwidth increase
Solution Approach 1:
The patent performs preliminary edge detection and gradient calculation before applying the sample adaptive offset filter. By pre-identifying edge locations and calculating gradient magnitudes in advance, the system can then apply the offset filter only where needed (at edges) rather than processing the entire image, significantly reducing computational complexity and memory access requirements.
Solution Approach 2:
Instead of applying the sample adaptive offset filter uniformly across the entire image (excessive action), the patent applies it partially only to regions where edges are detected and distortion is most likely to occur. This selective application maintains filtering effectiveness while dramatically reducing computational burden and memory access bandwidth.
3Manufacturing precision
If in-loop filtering is applied to minimize distortion, then picture quality improves, but processing time and computational resources increase
Solution Approach 1:
The patent segments the in-loop filtering into two distinct passes: a deblocking filter pass that operates on block boundaries, and a sample adaptive offset filter pass that operates on edge regions. This segmentation allows each filter to be optimized for its specific function and enables parallel processing of different image regions, reducing total processing time while maintaining quality improvement.
Solution Approach 2:
The patent applies filtering operations partially rather than uniformly across the entire image. By focusing computational resources only on block boundaries for deblocking and on detected edge regions for sample adaptive offset, the system achieves significant distortion reduction with substantially lower processing time and computational resource consumption.
Data Source
AI summary
The present invention relates to a video encoding/decoding method and apparatus. The video decoding method according to the present invention may comprise decoding filter information on a coding unit; classifying samples in the coding unit into classes on a per block classification unit basis; and filtering the coding unit having the samples classified into the classes on a per block classification unit basis by using the filter information.