Subpicture Group ALF Identifier Allocation for VVC
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing video coding standards, particularly Versatile Video Coding (VVC), face challenges in coordinating adaptive loop filtering (ALF) coefficients when encoding viewport-dependent omnidirectional video, leading to issues in aggregating bitstreams and merging subpictures from different quality and resolution versions into a single coded picture, especially in viewport-adaptive streaming applications.
Innovation Solution
The method involves dividing input pictures into subpictures, encoding them into versions with varying quality and/or resolution, partitioning these subpictures into groups, and allocating unique ranges of adaptive loop filter parameter set identifiers to each group, allowing for flexible allocation and merging of ALF parameters across subpicture groups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple subpicture versions with different quality and resolution are encoded independently, then viewport-adaptive streaming capability is improved, but coordination of adaptive loop filtering coefficients becomes problematic
Solution Approach 1:
The patent divides the picture into multiple subpictures and further into subpicture groups, with each group allocated a unique range of ALF parameter set identifiers. This segmentation allows independent encoding of multiple subpicture versions while maintaining coordinated filtering through structured identifier allocation, resolving the contradiction between adaptability and complexity.
Solution Approach 2:
The patent applies different ALF parameter sets to different subpicture groups based on their specific quality and resolution characteristics. By allocating unique identifier ranges to each subpicture group, the system enables localized optimization of filtering parameters tailored to each subpicture's properties, improving overall adaptability while managing complexity through systematic organization.
2Loss of energy
If subpictures are encoded at different quality and resolution levels, then bitrate optimization is improved, but merging of bitstreams becomes problematic
Solution Approach 1:
The patent pre-allocates unique ranges of ALF parameter set identifiers to each subpicture group before encoding. This preliminary organization establishes a coherent framework that guides the encoding process, ensuring that bitstreams from different quality and resolution levels can be systematically merged later without coordination problems.
Solution Approach 2:
The patent creates a universal identifier allocation scheme that works across all subpicture groups regardless of their specific quality or resolution characteristics. This universal approach enables any combination of subpicture versions to be merged into a single bitstream using the same coordination mechanism, improving ease of manufacture while maintaining bitrate efficiency.
3Manufacturing precision
If adaptive loop filtering is applied to each subpicture, then coding rate-distortion performance is improved, but parameter management complexity increases
Solution Approach 1:
The patent segments the parameter management task by assigning unique identifier ranges to each subpicture group. This segmentation allows the system to manage ALF parameters for each subpicture group independently while maintaining overall coherence, improving rate-distortion performance without overwhelming complexity in parameter management.
Solution Approach 2:
The patent systematically changes the parameter identification approach by introducing unique identifier ranges for different subpicture groups. This parameter reorganization enables efficient management of multiple ALF parameter sets by providing clear differentiation and selection mechanisms, improving coding performance while simplifying parameter management through structured identification.
Data Source
AI summary
A method comprising: dividing pictures of one or more input picture sequences into a plurality of subpictures (1000); encoding each of the subpictures into a plurality of subpicture versions having different quality and/or resolution (1002); partitioning the plurality of subpicture versions into one or more subpicture groups (1004); and allocating a range of adaptive loop filter parameter set identifiers coefficients for each of said subpicture groups (1006).