Mesh Subdivision Vertex Prediction for Non-Boundary Edges
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding systems face challenges in efficiently predicting the position of vertices in mesh subdivision, particularly when dealing with boundary and non-boundary edges, which affects compression efficiency and quality.
Innovation Solution
A coding device uses harmonic and arithmetic means to predict vertex positions based on the dot product of normal vectors and level of detail values, determining the appropriate mean based on edge type, and employs video-based dynamic mesh coding to transmit or receive meshes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single mean (arithmetic or harmonic) is used to predict vertex positions in mesh subdivision, then the method is simple and fast, but the prediction accuracy deteriorates when dealing with both boundary and non-boundary edges
Solution Approach 1:
The patent applies dynamics by making the prediction method adaptive rather than static. The system dynamically selects between arithmetic mean and harmonic mean based on the edge type (boundary or non-boundary) and surface curvature characteristics. This dynamic adaptation allows the system to optimize prediction accuracy for different geometric contexts without requiring a completely complex unified model.
Solution Approach 2:
The patent implements local quality by applying different prediction methods (arithmetic mean vs. harmonic mean) to different parts of the mesh based on local geometric properties. Boundary edges use one method while non-boundary edges use another, and within non-boundary edges, further differentiation is made based on surface curvature. This localized approach ensures each region gets the most appropriate prediction method for its specific characteristics.
2Measurement precision
If arithmetic mean is used for all vertex predictions, then the calculation is simple, but the prediction accuracy deteriorates for non-boundary edges with varying surface curvature
Solution Approach 1:
The patent changes the prediction parameter from a fixed arithmetic mean to a variable selection between arithmetic mean and harmonic mean based on surface curvature parameters. The system evaluates the dot product of normal vectors to determine surface curvature characteristics, then adjusts the prediction method accordingly. This parameter-based adaptation improves accuracy for non-boundary edges while maintaining reasonable computational complexity.
3Measurement precision
If harmonic mean is used for all vertex predictions, then the prediction accuracy improves for non-boundary edges, but the computational complexity and determination overhead increase
Solution Approach 1:
The patent segments the mesh edges into distinct categories (boundary edges and non-boundary edges) and further segments non-boundary edges based on surface curvature characteristics. This segmentation allows the system to apply the appropriate prediction method to each segment, avoiding the need to use the more complex harmonic mean for all edges. Only non-boundary edges with specific curvature properties trigger the harmonic mean calculation, reducing overall computational overhead.
4Measurement precision
If adaptive selection between arithmetic and harmonic mean is implemented, then the prediction accuracy improves, but the computational overhead and processing time increase
Solution Approach 1:
The patent applies partial action by selectively using the more computationally intensive harmonic mean only when necessary (for non-boundary edges with specific curvature characteristics), rather than applying it universally. For boundary edges and other cases, the simpler arithmetic mean suffices. This partial application of the complex method maintains prediction accuracy where needed while preserving processing speed in other regions.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
Systems, methods, and instrumentalities may be configured for mesh subdivision prediction. In examples, a coding device (e.g., an encoding and/or decoding device). A coding device may be configured to determine whether a mesh edge of a base mesh wherein a vertex is to be inserted is a non-boundary edge. Based on a determination that the mesh edge of the base mesh is a non-boundary edge, the device may use a harmonic mean to predict a position of the vertex.