Extended Coding Tree Units for 8K Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing AOMedia Video 1 (AV1) coding format, while expanding the partition-tree structure, still faces limitations in efficiency for large resolution videos such as 8K or 16K due to its maximum block size of 128×128 pixels, which restricts coding gains for larger image sizes.
Innovation Solution
The method involves dividing video frame data into coding tree units larger than 128×128 pixels and further subdividing them into smaller units for efficient encoding, using techniques like quadtree or binary tree splits, and limiting tools and prediction modes for larger units to enhance coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the maximum block size is limited to 128×128 pixels in AV1 format, then the partition-tree structure can be maintained with manageable complexity, but coding efficiency for large resolution videos (8K or 16K) is restricted
Solution Approach 1:
The patent divides video frames into multiple top-level coding tree units, each exceeding 128×128 pixels, and further segments these large units into smaller sub-blocks using quadtree or binary tree structures. This segmentation allows the system to process large resolution videos (8K or 16K) by breaking them into manageable pieces while maintaining the ability to handle fine-grained details through recursive subdivision.
2Manufacturing precision
If coding tree units are subdivided into smaller units, then encoding precision is improved, but the complexity of the encoding process increases
Solution Approach 1:
The patent implements dynamic subdivision of coding tree units based on content characteristics. The encoder recursively divides large coding tree units into smaller sub-blocks only where necessary, using adaptive quadtree or binary tree structures. This dynamic approach allows the system to achieve high encoding precision for complex regions while avoiding unnecessary subdivision in simpler areas, thereby controlling overall process complexity.
3Productivity
If larger coding units are used for high resolution videos, then compression efficiency improves, but the ability to capture fine details decreases
Solution Approach 1:
The patent employs multi-level segmentation where video frames are first divided into large top-level coding tree units for efficient compression, which are then recursively subdivided into smaller sub-blocks to preserve fine details. This hierarchical segmentation strategy enables the system to achieve both high compression efficiency through large unit processing and high detail capture through recursive subdivision.
Solution Approach 2:
The patent implements a nested structure of coding units where large top-level coding tree units contain smaller sub-blocks, which may themselves contain even smaller subdivisions. This nested organization allows the system to maintain multiple levels of detail representation within a unified coding framework, enabling both efficient compression at the coarse level and precise detail capture at finer levels.
Data Source
AI summary
A method, computer program, and computer system is provided for encoding video data. Data corresponding to a video frame is received. The video frame data is divided into one or more coding tree units having a height value greater than 128 pixels and a width value greater than 128 pixels. One or more of the coding tree units are further subdivided to a size smaller than 128 pixels by 128 pixels. The video data is encoded based on the divided and subdivided coding tree units.