An Algorithm to Triangulate Surfaces in 3D Using Unorganised Point Clouds
Reconstructing surfaces from a set of unorganised sample points in the 3D space is a very important problem in reverse engineering. Most algorithms first build a triangular mesh to obtain an approximate surface representation. In this paper we describe an algorithm which works by creating and merging local triangular complexes to obtain an unambiguous 2D-manifold triangulation. We use all the given sample points as vertices, which is a natural requirement. Our method is able to handle open boundaries and holes, different geni (for example tori) and unoriented surfaces in a computationally efficient way.
KeywordsPoint Cloud Delaunay Triangulation Triangular Mesh Neighbourhood Graph Triangle Mesh
Unable to display preview. Download preview PDF.
- Amenta, N., Bern, M., Kamvysselis, M.: A New Voronoi-based surface reconstruction algorithm. Comput. Graphics 415–421 (1998).Google Scholar
- Bajaj C., Bernardini, F., Chen, J., Schikore, D.: Automatic reconstruction of 3D CAD models. Proc. of the Int. J. Conf. on Theory and Practice of Geometric Modeling, Blaubeuren, Germany, October 1996.Google Scholar
- Bernardini, F.: Automatic reconstruction of CAD Models and properties from digital scans. Ph.D. Thesis, Purdue University, 1997.Google Scholar
- Edelsbrunner, H.: Weighted alpha shapes. Technical Report UIUCDCD-R- 92–1760. Comp. Sci. Dept., Univ. Illinois, Urbana, Ill, 1992.Google Scholar
- Guo, B., Mennon, J., Willette, B.: Surface reconstruction using alpha shapes. Comput. Graphics 16 177–190 (1997).Google Scholar
- Hoppe, H., et al.: Surface reconstruction from unorganised points. Comput. Graphics, 71–76 (1992).Google Scholar