An ultrametric approach to object recognition

Abstract : This paper presents a Bayes classifier with a hierarchical structure for appearance-based object recognition. It consists of a new kernel method, Ultrametric Spin Glass-Markov Random Fields, that integrates results of statistical physics with Gibbs distributions. Experiments show the effectiveness of our approach.
Type de document :
Communication dans un congrès
G. Greiner and H. Niemann and T. Ertl and B. Girod and H.-P. Seidel. Vision, Modeling, and Visualization (VMV '02), Nov 2002, Erlangen, Germany. 2002
Liste complète des métadonnées

https://hal.inria.fr/inria-00548260
Contributeur : Thoth Team <>
Soumis le : lundi 20 décembre 2010 - 08:42:29
Dernière modification le : mardi 21 décembre 2010 - 10:42:24

Identifiants

  • HAL Id : inria-00548260, version 1

Citation

Barbara Caputo, Gyuri Dorkó, Heinrich Niemann. An ultrametric approach to object recognition. G. Greiner and H. Niemann and T. Ertl and B. Girod and H.-P. Seidel. Vision, Modeling, and Visualization (VMV '02), Nov 2002, Erlangen, Germany. 2002. 〈inria-00548260〉

Partager

Métriques

Consultations de la notice

25