QTBT Block Partitioning for Higher-Efficiency JVET Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards like HEVC lack flexibility in block partitioning, leading to suboptimal coding efficiency for higher resolutions and bit-rates.
Innovation Solution
Implement a quadtree plus binary tree (QTBT) structure for JVET video coding, allowing for symmetric and asymmetric binary partitioning of coding units (CUs) to better match video data characteristics, with syntax elements to signal the partitioning structure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If HEVC block-based partitioning is used, then coding efficiency is maintained for standard resolutions, but flexibility is insufficient for higher resolutions and bit-rates
Solution Approach 1:
The patent applies segmentation by dividing the video block into a hierarchical structure of coding tree units (CTUs) that can be recursively partitioned into smaller coding units (CUs) using quadtree and binary tree structures. This multi-level segmentation enables flexible adaptation to different resolution requirements while maintaining coding efficiency through selective partitioning based on content characteristics.
Solution Approach 2:
The patent implements dynamic block partitioning where the partitioning structure is not fixed but adapts based on video content characteristics, resolution requirements, and bit-rate constraints. The encoder can dynamically choose between quadtree partitioning, binary tree partitioning, or no partitioning at all, allowing the system to optimize performance for each specific coding scenario rather than using a static partitioning scheme.
2Productivity
If more flexible block partitioning is implemented, then coding efficiency improves for higher resolutions, but syntax complexity and signaling overhead increase
Solution Approach 1:
The patent resolves the complexity issue by organizing the partitioning syntax into a hierarchical tree structure with multiple dimensions: the quadtree dimension for coarse partitioning and the binary tree dimension for fine-grained partitioning. This dimensional organization allows the encoder to signal partitioning information efficiently by only transmitting syntax elements for the actual partitioning decisions made, rather than encoding all possible partitioning parameters, thus reducing overall syntax complexity while maintaining flexibility.
3Adaptability or versatility
If asymmetric binary partitioning is added, then adaptability to video data characteristics improves, but overhead signaling increases
Solution Approach 1:
The patent applies partial action by implementing asymmetric binary partitioning selectively rather than universally. The syntax structure allows the encoder to choose whether to apply symmetric or asymmetric partitioning based on the local video content characteristics. Asymmetric partitioning is only activated when it provides actual coding benefit, such as when the video content has directional features that benefit from unequal block division, thereby avoiding unnecessary signaling overhead while maintaining adaptability where needed.
Data Source
AI summary
A method of partitioning in video coding for JVET, comprising representing a JVET coding tree unit as a root node in a quadtree plus binary tree (QTBT structure that can have quadtree or binary partitioning of the root node and quadtree or binary trees branching from each of the leaf nodes. The partitioning at any depth can use asymmetric binary partitioning to split a child node represented by a leaf node into two child coding units of unequal size, representing the two child coding units as leaf nodes in a binary tree branching from the parent leaf node and coding the child coding units represented by final leaf nodes of the binary tree with JVET. Disclosed is a generalized method of partitioning a block, either square or rectangular, which leads to more flexible block sizes with possible higher coding efficiency.