This work deals with the problem of efficient collision detection for point-based and polygon-based models .We have investigated the following three main areas: 1. We propose a technique that allows for increasing the speed of collusion detection by decreasing its quality. The main idea of our average-case approach is to estimate the probability of an intersection of two sets of polygons (or other primitives) in order to guide the simultaneous traversal of two bounding volume hierarchies. 2. We give a new definition of an implicit surface over a noisy point cloud based on the weighted least squares approach. We use a different kernel function that approximates geodesic distances by shortest paths in a geometric proximity graph. Thus, artefacts are reduced and the bandwidth and boundaries can automatically be determined. 3. We propose a point cloud hierarchy for collision detection between point clouds. This hierarchy approximates the surface at inner nodes and allows for excluding parts where no collision can occur. Moreover, our interpolation search can be used to find intersection points efficiently by utilizing the proximity graph.
Efficient collision detection for point and polygon based models
Effiziente Kollisionserkennung für Punkt-und Polygonmodelle
HNI-Verlagsschriftenreihe, Universität Paderborn ; 186 ; 1-153
2006
153 Seiten, Bilder, Tabellen, Quellen
Hochschulschrift
Deutsch
British Library Conference Proceedings | 2005
|IEEE | 2005
|British Library Conference Proceedings | 2016
|