Geometry Image Matching for Similarity Estimation of 3D Shapes

Authors:
Hamid Laga;Hiroki Takahashi;Masayuki Nakajima
Affiliations:
Tokyo Institute of Technology;Tokyo Institute of Technology;Tokyo Institute of Technology
Venue:
CGI '04 Proceedings of the Computer Graphics International
Year:
2004

Citing 0
Cited 3

Intraclass Retrieval of Nonrigid 3D Objects: Application to Face Recognition

IEEE Transactions on Pattern Analysis and Machine Intelligence
Multishape-features and text-feature integration on 3D model similarity retrieval

International Journal of Innovative Computing and Applications
Multi-feature integration with relevance feedback on 3D model similarity retrieval

Journal of Mobile Multimedia

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we describe our preliminary findings inapplying the spherical parametrization and geometry imagesto the task of 3D shape matching and similarity basedcomparison of polygon soup models. Unlike traditional approachwhere multiple 2D views of the same object are requiredto capture the relevant geometry features, our proposedtechnique uses spherical parametrization and geometryimages for 3D shape matching. This technique reducesthe problem to the 2D space without computing 2Dprojections and guaranties the preservation of small details.Moreover, we take advantage from the hierarchicalnature of the parametrization process, through the ProgressiveMesh simplification, to derive a multiresolution analysistechnique. Our proposed algorithm is invariant to similaritytransformations such as rotation and scaling. Theefficiency of this approach is discussed through a set of experiments.