Visual Mesh Skinning via Cluster Orientation Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing skinning techniques in computer graphics struggle to achieve smooth deformation of deformable objects, particularly in large plastic deformation and topological changes, as they either require a large number of clusters or restrict the visual mesh vertex to be within the convex hull of particles, which limits rotation reproduction and computational efficiency.
Innovation Solution
A method using position-based dynamics (PBD) that interpolates orientation from clusters to particles, allowing the visual mesh to be skinned without the need for the vertex to be within the convex hull, while maintaining rotation information and efficiently simulating deformable solids with large plastic deformation and topological changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If skinning is performed to clusters using shape matching techniques, then orientation information is utilized and visual vertex does not need to be within convex hull, but a very large number of clusters are required to achieve smooth deformation
Solution Approach 1:
The patent combines particle-based position data with cluster-based orientation data into a unified skinning approach. Each visual vertex associates with both particles (for position interpolation) and clusters (for orientation interpolation), merging the advantages of both representations to reduce the total number of computational elements needed while maintaining smooth deformation and rotation accuracy.
Solution Approach 2:
The skinning method uses a composite approach by interpolating position from particles and orientation from clusters simultaneously. This composite technique allows the system to achieve accurate visual deformation with fewer clusters than traditional cluster-only methods, as the particle position data compensates for the reduced cluster density.
2Quantity of substance
If skinning is performed to particles with dense sampling, then only a relatively small number of particles are required, but visual mesh vertex must be within convex hull of particles to prevent extrapolation artifacts
Solution Approach 1:
The patent applies asymmetric treatment to position and orientation data: position interpolation uses particle information within the convex hull, while orientation interpolation uses cluster information that can influence vertices outside the convex hull. This asymmetry allows visual mesh vertices to be positioned more freely without causing extrapolation artifacts in position, while still maintaining rotation accuracy through cluster-based orientation data.
3Loss of information
If displacement is interpolated in particle-based skinning, then rotation information may not be reproduced, but maintaining rotation information requires more complex computation
Solution Approach 1:
The patent segments the skinning computation into two independent parts: position interpolation from particles and orientation interpolation from clusters. This segmentation allows each part to be optimized separately - position interpolation remains computationally simple while orientation interpolation captures rotation information, avoiding the need for complex unified computations that would be required to preserve both simultaneously in a single system.
Data Source
AI summary
Embodiments of the present invention provide a method for simulating deformable solids undergoing large plastic deformation and topological changes using shape matching. Positional information for particles and orientation information from clusters is used to simulate deformable solids represented by particles. Each visual vertex stores references to particles that influence the vertex, and stores the local position of the particles. A two-step method interpolates orientation from clusters to particles, and uses the orientation and position of particles to skin the visual mesh vertices. This results in a fast method that can reproduce rotation and does not require the visual mesh vertex to be located within a convex hull of particles.