Adaptive Loop Filtering for High-Dynamic-Range Video
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards, such as HEVC, lack effective adaptive loop filtering techniques for high-dynamic-range (HDR) video, which limits coding efficiency and image quality, especially in high-resolution content.
Innovation Solution
The development of an adaptive loop filtering method that classifies input blocks based on luminance information and adjusts filter coefficients for HDR content, using techniques like luminance-based classification, gradient weighting, and separate thresholds for HDR content, to enhance coding efficiency and image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional video coding standards (HEVC) are used without HDR-specific adaptive loop filtering, then device complexity is reduced and ease of operation is maintained, but coding efficiency and image quality deteriorate for HDR content
Solution Approach 1:
The patent segments the filtering process by classifying blocks into HDR and SDR types based on luminance values. Different filtering modes are applied to different block types: ALF for HDR blocks and traditional loop filters for SDR blocks. This segmentation allows optimized filtering for HDR content while maintaining compatibility with existing standards for standard content.
Solution Approach 2:
The patent introduces dynamic adaptation of filtering parameters based on block characteristics. The filter decision flag and luminance class values enable the system to dynamically select appropriate filtering modes (ALF or traditional) for each block, optimizing performance for varying content types and luminance ranges.
2Manufacturing precision
If HDR-specific adaptive loop filtering with luminance-based classification is implemented, then image quality and coding efficiency improve for HDR content, but device complexity and processing overhead increase
Solution Approach 1:
The patent applies local quality optimization by using luminance-based classification to identify HDR blocks and applying ALF specifically to those regions. The luminance class value determines whether a block receives ALF processing, ensuring that enhanced filtering is applied locally where it is most beneficial rather than uniformly across all content.
Solution Approach 2:
The patent changes filtering parameters based on luminance characteristics. By computing luminance class values and using them to select filtering modes, the system adapts filter parameters (such as filter coefficients and application decisions) to match the local luminance properties of each block, optimizing image quality for HDR content.
3Productivity
If separate filtering processes are applied for HDR and SDR blocks, then coding efficiency improves and image quality is enhanced, but processing time and computational load increase
Solution Approach 1:
The patent performs preliminary classification of blocks into HDR and SDR categories using luminance-based criteria before applying filtering. By computing luminance class values in advance and making filter decisions based on these pre-computed values, the system avoids redundant processing and streamlines the filtering workflow, reducing overall processing time.
Data Source
Figure 1A
Figure 1B
Figure 2A
AI summary
Methods, processes, and systems are presented for adaptive loop filtering in coding and decoding high dynamic range (HDR) video. Given an input image block, its luminance information may be used to adapt one or more parameters of adaptive loop filtering and compute gradient and directionality information, activity information, a classification index, and adaptive-loop-filtering coefficients.