Login
TitleBicubic Subdivision-Surface wavelets for Large-Scale Isosurface Representation and Visualization (In Proceedings)
inIEEE Visualization 2000
Author(s) Martin Bertram, Mark A. Duchaineau, Bernd Hamann, Ken Joy
Editor(s) Thomas Ertl, Bernd Hamann, A. Varshney
Keyword(s)Compression Algorithms, Geometric Modeling, Isosurfaces, Multiresolution Methods, Wavelets
Year 2000
LocationSalt Lake City, Utah
DateOctober 9--13, 2000
PublisherIEEE Computer Society Press
AddressLos Alamitos, California
OrganizationIEEE
Pages389--396
Download
BibTeX
Abstract We introduce a new subdivision-surface wavelet transform for arbitrary two-manifolds with boundary that is the first to use simple lifting-style filtering operations with bicubic precision. We also describe a conversion process for re-mapping large-scale isosurfaces to have subdivision connectivity and fair parameterizations so that the new wavelet transform can be used for compression and visualization. The main idea enabling our wavelet transform is the circular symmetrization of the filters in irregular neighborhoods, which replaces the traditional separation of filters into two 1-D passes. Our wavelet transform uses polygonal base meshes to represent surface topology, from which a Catmull-Clark-style subdivision hierarchy is generated. The details between these levels of resolution are quickly computed and compactly stored as wavelet coefficients. The isosurface conversion process begins with a contour triangulation computed using conventional techniques, which we subsequently simplify with a variant edge-collapse procedure, followed by an edge-removal process. This provides a coarse initial base mesh, which is subsequently refined, relaxed and attracted in phases to converge to the contour. The conversion is designed to produce smooth, untangled and minimally- skewed parameterizations, which improves the subsequent compression after applying the transform. We have demonstrated our conversion and transform for an isosurface obtained from a high-resolution turbulent-mixing hydrodynamics simulation, showing the potential for compression and level-of-detail visualization.