High dimensional data clustering

Charles Bouveyron 1, 2 Stéphane Girard 2 Cordelia Schmid 1, *
* Auteur correspondant
1 LEAR - Learning and recognition in vision
GRAVIR - IMAG - Graphisme, Vision et Robotique, Inria Grenoble - Rhône-Alpes, CNRS - Centre National de la Recherche Scientifique : FR71
Abstract : Clustering in high-dimensional spaces is a recurrent problem in many domains, for example in object recognition. High-dimensional data usually live in different lowdimensional subspaces hidden in the original space. This paper presents a clustering approach which estimates the specific subspace and the intrinsic dimension of each class. Our approach adapts the Gaussian mixture model framework to high-dimensional data and estimates the parameters which best fit the data. We obtain a robust clustering method called High- Dimensional Data Clustering (HDDC). We apply HDDC to locate objects in natural images in a probabilistic framework. Experiments on a recently proposed database demonstrate the effectiveness of our clustering method for category localization.
Type de document :
Communication dans un congrès
Alfredo Rizzi and Maurizio Vichi. 17th International Conference on Computational Statistics (Compstat '06), Aug 2006, Rome, Italy. Springer-Verlag, pp.813--820, 2006
Liste complète des métadonnées

Littérature citée [10 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00548591
Contributeur : Thoth Team <>
Soumis le : jeudi 6 janvier 2011 - 09:21:10
Dernière modification le : mercredi 11 avril 2018 - 01:56:15
Document(s) archivé(s) le : jeudi 7 avril 2011 - 02:30:38

Fichier

paper_COMPSTAT06.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00548591, version 1

Collections

IMAG | INRIA | UGA

Citation

Charles Bouveyron, Stéphane Girard, Cordelia Schmid. High dimensional data clustering. Alfredo Rizzi and Maurizio Vichi. 17th International Conference on Computational Statistics (Compstat '06), Aug 2006, Rome, Italy. Springer-Verlag, pp.813--820, 2006. 〈inria-00548591〉

Partager

Métriques

Consultations de la notice

418

Téléchargements de fichiers

453