Video Block Partition Signalling with Predefined-Partition Shortcuts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing video coding standards, such as VVC, face inefficiencies in block partitioning due to excessive syntax overhead and complex decision-making processes for determining partition modes, which affect coding efficiency and compression performance.
Innovation Solution
The proposed method introduces a shortcut mode using predefined partitions and simplified syntax signaling mechanisms to quickly determine the partition mode for video blocks, reducing the need for multiple decision steps and minimizing syntax overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional block partitioning methods are used in video coding, then comprehensive partition coverage is achieved, but syntax overhead increases and decision-making complexity increases
Solution Approach 1:
The partitioning process is segmented into two distinct stages: a first stage that always applies a first partitioning type (e.g., quadtree), and a second stage that conditionally applies a second partitioning type (e.g., ternary tree or binary tree). This segmentation allows the system to reduce syntax overhead by making the second stage optional while still providing comprehensive partition coverage when needed.
Solution Approach 2:
The partitioning system dynamically adapts its behavior based on content characteristics. The decoder determines whether to apply the second partitioning type based on conditions such as block size, gradient calculations, or edge detection results. This dynamic adaptation reduces syntax overhead for simple content while maintaining partition versatility for complex content requiring multiple partitioning types.
2Adaptability or versatility
If multiple partitioning types are always applied, then partition versatility is improved, but processing complexity increases
Solution Approach 1:
The first partitioning type is applied preliminarily to all blocks before considering the second partitioning type. This preliminary action establishes a baseline partition structure, and only blocks that meet specific criteria (determined by gradient analysis, edge detection, or size thresholds) proceed to the second partitioning stage. This reduces decision-making complexity by filtering out blocks that don't require complex partitioning.
Solution Approach 2:
Different partitioning strategies are applied to different regions of the video content based on local characteristics. Blocks with high gradient values, detected edges, or specific size ranges trigger the application of the second partitioning type, while other blocks use only the first partitioning type. This local quality approach maintains partition versatility where needed while reducing processing complexity in uniform regions.
3Loss of information
If conditional partitioning is used, then syntax overhead is reduced, but coding efficiency may be compromised
Solution Approach 1:
The system incorporates feedback mechanisms where the decoder evaluates block characteristics (such as gradient magnitude, edge presence, or variance) and uses this feedback to determine whether applying the second partitioning type would improve coding efficiency. This feedback-driven approach ensures that the conditional application of partitioning types maintains coding efficiency by selectively applying complex partitioning only when beneficial.
Solution Approach 2:
The system changes parameters such as block size thresholds, gradient thresholds, or partitioning depth limits based on content characteristics and coding conditions. These parameter adjustments allow the conditional partitioning system to adapt to different video content types and quality requirements, maintaining coding efficiency across diverse scenarios while reducing syntax overhead for appropriate content.
Data Source
AI summary
Methods and apparatus for video coding. A current block is partitioned into one or more sub-blocks. The partitioning process comprises: signalling or parsing a first syntax where the first syntax is indicative of whether to split the current block; and in response to the first syntax indicating the current block being split, signalling or parsing a second syntax where the second syntax is indicative of whether to use a shortcut mode to indicate a target partition from predefined partitions directly. The sub-blocks are then encoded or decoded. According to another method, the partitioning process comprises signalling or parsing a first syntax where the first syntax is indicative of whether to split the current block; and in response to the first syntax, signalling or parsing a second syntax where the second syntax is indicative of one or more conditions comprising whether one of predefined partitions being applied to the current block.