Markov random fields for textures recognition with local invariant regions and their geometric relationships

Juliette Blanchet 1, 2 Florence Forbes 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 : This paper describes a new probabilistic framework for recognizing textures in images. Images are described by local affine-invariant descriptors and their spatial relationships. We introduce a statistical parametric models of the dependence between descriptors. We use Hidden Markov Models (HMM) and estimate the parameters with a recent technique based on the mean field principle. Preliminary results for texture recognition are promising and outperform existing techniques.
Type de document :
Communication dans un congrès
William Clocksin and Andrew Fitzgibbon and Philip Torr. British Machine Vision Conference (BMVC '05), Sep 2005, Oxford, United Kingdom. The British Machine Vision Association (BMVA), 2005, 〈http://www.bmva.org/bmvc/2005/papers/paper-57-179.html〉
Liste complète des métadonnées

Littérature citée [16 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00548520
Contributeur : Thoth Team <>
Soumis le : lundi 20 décembre 2010 - 09:08:58
Dernière modification le : mercredi 11 avril 2018 - 01:53:12
Document(s) archivé(s) le : lundi 21 mars 2011 - 03:08:28

Fichier

BFS05a.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00548520, version 1

Collections

IMAG | INRIA | UGA

Citation

Juliette Blanchet, Florence Forbes, Cordelia Schmid. Markov random fields for textures recognition with local invariant regions and their geometric relationships. William Clocksin and Andrew Fitzgibbon and Philip Torr. British Machine Vision Conference (BMVC '05), Sep 2005, Oxford, United Kingdom. The British Machine Vision Association (BMVA), 2005, 〈http://www.bmva.org/bmvc/2005/papers/paper-57-179.html〉. 〈inria-00548520〉

Partager

Métriques

Consultations de la notice

289

Téléchargements de fichiers

203