Adaptive Loop Filter Set Selection in Video Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video encoding and decoding technologies face challenges in selecting optimal filter sets for improving video quality without increasing signaling overhead, particularly in handling different types of video content.
Innovation Solution
A method for video encoders and decoders to select fixed filter sets based on a quantization parameter and additional characteristics of video blocks, such as temporal layer ID, transform type, or hash hit percentage, rather than relying solely on quantization parameters, to enhance filter selection for specific content types like natural or screen content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If filter sets are selected based solely on quantization parameter, then the selection process is simple and fast, but the filter selection is not well-suited for specific types of video content
Solution Approach 1:
The patent applies local quality by selecting different filter sets based on the specific characteristics of each video block (natural content vs. screen content) rather than using a uniform filtering approach. The decoder identifies block characteristics and applies appropriate filters from different sets, making the filtering adaptive to local content requirements.
Solution Approach 2:
The patent changes the parameter basis for filter selection from solely quantization parameter to include additional characteristics such as block type (inter, intra, intra copy, palette). This parameter expansion enables more accurate filter selection while maintaining a systematic approach that doesn't significantly increase complexity.
2Measurement precision
If multiple characteristics are used for filter set selection, then filter selection becomes more suitable for specific video content, but signaling overhead increases
Solution Approach 1:
The patent applies self-service by deriving block characteristics from existing decoded video data and previously decoded information rather than requiring additional signaling. The decoder independently determines block type and other characteristics from the video stream and reconstruction data, eliminating the need for extra signaling overhead.
3Productivity
If fixed filter sets are selected without considering block characteristics, then processing is faster and simpler, but decoded video quality is reduced
Solution Approach 1:
The patent applies segmentation by dividing video blocks into different categories (inter blocks, intra blocks, intra copy blocks, palette prediction blocks) and applying appropriate filter sets to each segment. This segmentation enables targeted filtering that improves quality while maintaining efficient processing through classification-based organization.
Solution Approach 2:
The patent applies preliminary action by pre-defining multiple fixed filter sets with different characteristics before decoding begins. The decoder can then quickly select from these pre-prepared filter sets based on block characteristics, avoiding the need for complex real-time filter design while still achieving content-adaptive filtering.
Data Source
AI summary
A device for decoding video data, as part of applying a first stage of an adaptive loop filter (ALF) to a reconstructed sample of a reconstructed block for a color component, selects a first set of filters from a plurality of sets of fixed filters based on a quantization parameter of the block of video data and a second characteristic of the block of video data; determines a first class index for the reconstructed sample; selects a filter from a first set of filters based on the first class index; and applies the filter from the first set of filters to the reconstructed sample to determine a first intermediate sample value.