Geometry synthesis
Ares Lagae | Olivier Dumont | Philip Dutré |
Contact: Ares Lagae
Report CW 381, Departement Computerwetenschappen, Katholieke Universiteit Leuven, Celestijnenlaan 200A, 3001 Heverlee, Belgium, March 2004
Abstract
Inspired by texture synthesis techniques, we present in this paper a method for geometry synthesis. Given an example of input geometry, we synthesize new output geometry that is perceived similar to the input geometry, but at the same time differs in its local appearance. We assume our input geometry satisï¬es the constraints of a Markov Random Field model, and represent the input by a hierarchical distance ï¬eld. This allows us to perform fast matching queries between a target distance ï¬eld that is partially synthesized, and the input distance ï¬eld. Once the target distance ï¬eld is completed, we copy the original corresponding geometry elements to our synthesized result. We show that automatically generating geometry by example can be achieved within reasonable computing times, and is able to produce convincing results.