Fast Discrete Intersection Detection for Cloth Penetration Resolution

Juntao Ye,Timo R. Nyberg,Gang Xiong
DOI: https://doi.org/10.1109/BigMM.2015.22
2015-01-01
Abstract:Penetrations are often unavoidable in modeling cloth or other deformable surfaces. It could exist as an initial configuration or show up in the middle of a dynamics simulation process. We present a new method for resolving such penetrations. To be adapted to a wide range of applications, this method is based on history-free discrete intersection detection (DID). It is also orientation-free as it does not assume any front- and back-face identification. Our method relies on dynamic repulsive normal (DRN) to compute proper displacements to relocate vertices to be intersection-free. First, intersection contours are constructed and classified, followed by a global analysis of the collision configurations. Then for some types of configurations, penetrating (illegal) regions are identified using a heuristic paradigm "small region is illegal". For those surfaces having identifiable illegal regions, proper displacements for incorrectly configured vertices are computed using DRN. For those configurations that do not clearly define legal/illegal regions, displacements are designed to push one mesh to the boundary of the other. The proposed method can also be used in the context of time-dependent simulation of complex deformable surfaces, making it an competitive alternative to the popular CCD-based approach.
What problem does this paper attempt to address?