Submitted for publication, January 2001 Consistent Mesh Parameterizations Emil Praun

Figure 1: When given a set of head models an obvious shape to compute is their average. In general the connectivity and sampling pattern of the models is different and computing the average is non trivial. After computing consistent mesh parameterizations (red patch boundaries) and remeshing, all models have the same connectivity and sampling pattern so computing the average becomes trivial. A basic element of Digital Geometry Processing algorithms is the establishment of a smooth parameterization for a given model. In this paper we propose an algorithm which establishes parameteriza-tions for a set of models. The parameterizations are called consis-tent because they share the same base domain and respect features. They give immediate correspondences between models and allow remeshes with the same connectivity. Such remeshes form the ba-sis for a large class of algorithms, including principal component analysis, wavelet transforms, detail and texture transfer between models, and n-way shape blending. We demonstrate the versatil-ity of our algorithm with a number of examples.