Block-Based Adaptive Loop Filter Design for Video Coding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video coding technologies face inefficiencies in adapting filtering techniques to local video data characteristics, leading to suboptimal coding efficiency and picture quality.

Innovation Solution

A method where a video encoder and decoder determine sets of adaptive loop filters on a per-block basis, allowing the selection of filters tailored to the local characteristics of the video data, improving coding efficiency and picture quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a single set of adaptive loop filters is used for the entire picture, then device complexity and signaling overhead are reduced, but coding efficiency and picture quality deteriorate due to inability to adapt to local characteristics

Engineering Contradiction:
Improveadaptability to local video data characteristicsVSAvoidcomplexity of filter selection and signaling
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The picture is divided into multiple blocks, and each block can have its own set of adaptive loop filters selected independently. This segmentation allows the filtering process to adapt to local characteristics of different regions while maintaining manageable complexity through block-level independence.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different sets of adaptive loop filters are applied to different blocks based on their local characteristics. Each block's filter selection is tailored to its specific content requirements, enabling localized optimization of picture quality and coding efficiency without requiring uniform filtering across the entire picture.

Inventive Principle:
Principle #3Local quality

2Productivity

If multiple sets of adaptive loop filters are selected per block, then coding efficiency and picture quality improve through better local adaptation, but signaling overhead and processing complexity increase

Engineering Contradiction:
Improvecoding efficiencyVSAvoidsignaling overhead
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

Multiple sets of candidate adaptive loop filters are prepared in advance for each block. The encoder selects the most appropriate set from these pre-prepared candidates based on local characteristics, and this selection is signaled to the decoder. This preliminary preparation enables efficient coding while controlling signaling overhead through selective transmission of only necessary selection information.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If adaptive loop filtering is applied to all blocks, then picture quality improves through consistent filtering, but processing time and computational complexity increase

Engineering Contradiction:
Improvepicture qualityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

Adaptive loop filtering is applied selectively to blocks where it provides the most benefit, rather than uniformly to all blocks. The encoder evaluates local characteristics and applies filtering only where needed, achieving good picture quality in critical regions while reducing overall processing time and computational complexity through targeted application.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3804337B1Block-based adaptive loop filter (ALF) design and signaling
Publication Date: 2024.11.27 QUALCOMM INC
  • EP3804337B1 patent drawingFigure 1
  • EP3804337B1 patent drawingFigure 2
  • EP3804337B1 patent drawingFigure 3A~3D

AI summary

A video encoder and video decoder may determine a set of adaptive loop filters, from among a plurality of sets of adaptive loop filters, on a per-block basis. Each set of adaptive loop filters may include filters from a previous picture, filters signaled for the current picture, and/or pre-trained filter. By varying the set of adaptive loop filters on a per-block basis, the adaptive loop filters available for each block of video data may be more adapted to local statistics of the video data.