Tile and Subpicture Partitioning for Independent Video Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards face challenges in efficiently decoding and encoding video streams with tile and sub-picture partitioning, particularly in scenarios requiring independent decoding of sub-pictures and avoiding processing across tile boundaries.
Innovation Solution
Implementing a method and system that allows for independent decoding of sub-pictures within a video stream by using tile and sub-picture partitioning, with flags indicating whether processing and prediction across tile boundaries are allowed, and enabling in-loop filtering operations only within sub-picture boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If tile and sub-picture partitioning is implemented for independent decoding, then decoding efficiency and parallel processing capability are improved, but device complexity and processing difficulty increase
Solution Approach 1:
The video picture is divided into multiple independent tiles and sub-pictures that can be decoded in parallel. Each tile boundary is marked with flags (first flag and second flag) that control processing independence, allowing the decoder to process multiple regions simultaneously while maintaining the option to perform in-loop filtering across boundaries when needed.
2Productivity
If processing and prediction across tile boundaries is disallowed, then parallel processing capability is improved, but video quality and compression efficiency deteriorate
Solution Approach 1:
The system dynamically controls processing across tile boundaries using two flags: the first flag indicates whether any processing and prediction across tile boundaries is disallowed, and the second flag indicates whether in-loop filtering operations may be performed across tile boundaries. This allows adaptive adjustment of parallel processing vs. quality trade-offs based on content characteristics.
3Reliability
If in-loop filtering operations are performed across tile boundaries, then video quality is improved, but decoding complexity and processing time increase
Solution Approach 1:
In-loop filtering operations are selectively applied across tile boundaries based on the second flag value. When the second flag indicates filtering is allowed, the decoder performs in-loop filtering across boundaries to improve quality in regions that benefit from it, while maintaining independent processing elsewhere to reduce overall complexity.
4Adaptability or versatility
If flags for controlling processing across tile boundaries are added, then flexibility and adaptability are improved, but data overhead and signaling complexity increase
Solution Approach 1:
The system uses two binary flags (first flag and second flag) to control processing behavior across tile boundaries. These compact parameter representations provide multiple processing modes (fully independent, filtering allowed, prediction allowed) with minimal data overhead compared to more complex signaling schemes.
Data Source
AI summary
Systems and methods for decoding are provided, a method includes: encoding a first sub-picture of a picture, independently from a second sub-picture of the picture, using sub-picture, tile group, and tile partitioning: and sending, to at least one decoder, at least one coded sub-bitstream of a coded video stream that includes the first sub-picture and the second sub-picture, wherein (i) the first sub-picture is a first rectangular region of the picture and the second sub-picture is a second rectangular region of the picture, the second rectangular region being different from the first rectangular region, (ii) the first sub-picture includes a first plurality of tiles, and the second sub-picture includes at least one tile, (iii) the first sub-picture and the second sub-picture do not share a common tile, and (iv) the first plurality of tiles of the first sub-picture are grouped into at least two first tile groups.