Mesh Segmentation Using Laplacian Eigenvectors and Gaussian Mixtures

Avinash Sharma 1 Radu Horaud 1 David Knossow 1 Etienne von Lavante 1
1 PERCEPTION - Interpretation and Modelling of Images and Videos
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : In this paper a new completely unsupervised mesh segmentation algorithm is proposed, which is based on the PCA interpretation of the Laplacian eigenvectors of the mesh and on parametric clustering using Gaussian mixtures. We analyse the geometric properties of these vectors and we devise a practical method that combines single-vector analysis with multiple-vector analysis. We attempt to characterize the projection of the graph onto each one of its eigenvectors based on PCA properties of the eigenvectors. We devise an unsupervised probabilistic method, based on one-dimensional Gaussian mixture modeling with model selection, to reveal the structure of each eigenvector. Based on this structure, we select a subset of eigenvectors among the set of the smallest non-null eigenvectors and we embed the mesh into the isometric space spanned by this selection of eigenvectors. The final clustering is performed via unsupervised classification based on learning a multi-dimensional Gaussian mixture model of the embedded graph.
Document type :
Conference papers
Complete list of metadatas


https://hal.inria.fr/inria-00446990
Contributor : Radu Horaud <>
Submitted on : Wednesday, January 13, 2010 - 10:45:47 PM
Last modification on : Wednesday, April 11, 2018 - 1:59:37 AM
Long-term archiving on : Thursday, October 18, 2012 - 12:30:31 PM

Files

SharmaAAAIFall2009.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00446990, version 1

Collections

Citation

Avinash Sharma, Radu Horaud, David Knossow, Etienne von Lavante. Mesh Segmentation Using Laplacian Eigenvectors and Gaussian Mixtures. AAAI Fall Symposium on Manifold Learning and its Applications, AAAI, Nov 2009, Arlington, VA, United States. pp.50-56. ⟨inria-00446990⟩

Share

Metrics

Record views

557

Files downloads

426