Geometric Partition Refinement for Screen Content Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face inefficiencies in partitioning and predicting video sequences, particularly for screen content and complex scenes, leading to suboptimal compression and decoding performance.
Innovation Solution
Implement geometric partitioning refinement techniques, including quadtree+multi-type tree partitioning and advanced intra prediction modes, to enhance block partitioning and prediction accuracy, especially for screen content and complex scenes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional partitioning methods are used, then device complexity is reduced, but video coding efficiency and compression performance deteriorate
Solution Approach 1:
The patent applies segmentation by dividing the video block into multiple partition types (symmetrical and asymmetrical partitions) using different tree structures (binary, ternary, quadtree). This allows the encoder to select the most appropriate partitioning pattern for different regions of the video content, improving coding efficiency by adapting to local characteristics while managing complexity through structured division strategies.
Solution Approach 2:
The patent implements dynamic partitioning where the partitioning structure is not fixed but adapts based on the video content characteristics. The encoder can dynamically switch between different partition types (symmetrical/asymmetrical) and tree structures based on rate-distortion optimization, allowing the system to respond to varying scene complexity and content requirements rather than using a static partitioning approach.
2Measurement precision
If simple partitioning is used, then ease of operation is improved, but prediction accuracy for screen content and complex scenes deteriorates
Solution Approach 1:
The patent applies local quality by allowing different partitioning strategies and prediction modes to be applied to different regions of the video block. Symmetrical partitions may be used for uniform regions while asymmetrical partitions handle regions with directional features or edges. This localized adaptation of partitioning quality improves prediction accuracy for specific content types without requiring complex partitioning throughout the entire block.
Solution Approach 2:
The patent introduces asymmetry by providing both symmetrical and asymmetrical partition options. Asymmetrical partitions allow unequal division of blocks, which is particularly useful for screen content with text lines, graphical edges, or regions where one dimension has different characteristics than the other. This asymmetric capability enhances prediction accuracy for non-uniform content while maintaining symmetrical options for simpler regions.
3Productivity
If advanced partitioning techniques are implemented, then compression performance improves, but processing time increases
Solution Approach 1:
The patent applies partial action by implementing a hierarchical partitioning approach where not all blocks require the most complex partitioning analysis. The encoder can perform rapid evaluation of partition options and apply full asymmetric/ternary analysis only where needed, rather than exhaustively evaluating all partition types for every block. This selective application reduces processing time while maintaining compression performance where it matters most.
Solution Approach 2:
The patent utilizes parameter changes by allowing dynamic adjustment of partitioning parameters such as partition depth, tree type selection, and symmetry constraints based on content characteristics and rate-distortion tradeoffs. By changing these parameters adaptively rather than using fixed settings, the system optimizes compression performance for different content types while managing processing complexity through parameter control rather than structural complexity.
Data Source
AI summary
A video coder (encoder or decoder) determines a partition mode that indicates an initial partition edge for partitioning a current block. The video coder identifies reference samples neighboring the current block based on the initial partition edge. The video coder determines a refined partition edge based on the reference samples. The video coder generates a prediction of the current block by predicting at least one part of the current block defined by the refined partition edge. The video coder generates a reconstruction of the current block based on the prediction of the current block.