Geometric Partition Mode Displacement Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Draft video standards for geometric partition mode (GEO) often result in larger residuals and worse rate-distortion values, and have a large number of available GEO modes that increase memory usage, processing time, and overhead.
Innovation Solution
Improving displacement values for certain GEO angles to reduce residuals and simplifying storage by reducing the number of available GEO modes, which decreases memory usage and processing complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a large number of GEO modes are available in video coding standards, then the adaptability and versatility of the coding system is improved, but the device complexity, memory usage, and processing time increase
Solution Approach 1:
The patent segments the GEO mode selection process by dividing the prediction unit into multiple sub-blocks and applying different GEO modes to different sub-blocks based on local characteristics. This allows the system to use fewer overall modes while achieving adaptability through localized differentiation, reducing memory usage and processing complexity while maintaining versatility.
Solution Approach 2:
The patent applies different GEO mode parameters (such as different displacement values or angle configurations) to different spatial regions within a prediction unit based on local image characteristics. This local quality approach allows the coding system to achieve high adaptability in specific regions without requiring a large number of global modes, thereby reducing overall device complexity and memory requirements.
2Manufacturing precision
If displacement values are optimized for certain GEO angles, then the manufacturing precision and measurement precision are improved, but the device complexity increases
Solution Approach 1:
The patent optimizes displacement values for specific GEO angles (such as 45 degrees) by adjusting the parameter relationships between distance index and displacement value. For example, setting the displacement value to half of what would be expected for a given distance index creates more accurate split locations for common angles while maintaining a simple parameter structure that does not significantly increase device complexity.
3Productivity
If the number of GEO modes is reduced, then the device complexity and processing time are reduced, but the adaptability and versatility decrease
Solution Approach 1:
The patent compensates for reducing the number of overall GEO modes by segmenting the prediction unit into multiple sub-blocks and applying different modes to different segments. This segmentation allows the system to process fewer modes overall (improving productivity) while still achieving diverse adaptation through localized mode application (maintaining versatility).
Solution Approach 2:
The patent introduces dynamic selection of GEO modes based on local image characteristics and prediction unit properties. This dynamic approach allows a smaller set of modes to be adaptively applied where needed, improving processing speed by avoiding exhaustive mode searching while maintaining versatility through context-dependent mode selection.
Data Source
AI summary
An example device for decoding video data includes a memory configured to store the video data and one or more processors implemented in circuitry and coupled to the memory. The one or more processors are configured to determine a first distance index associated with a first geometric partition mode (GEO) angle for a first prediction unit (PU) of the video data to be 4. The one or more processors are configured to determine a first displacement value based on the first distance index, the first displacement value being indicative of a distance from a center of the first PU to a GEO split. The one or more processors are configured to decode the first PU based on the first GEO angle and the first displacement value. The first displacement value is half of a displacement value associated with a distance index of 2.