Interpretable Visual Models for Human Perception-Based Object Retrieval

Ahmed Rebai 1 Alexis Joly 1, 2 Nozha Boujemaa 1
2 ZENITH - Scientific Data Management
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier, CRISAM - Inria Sophia Antipolis - Méditerranée
Abstract : Understanding the results returned by automatic visual concept detectors is often a tricky task making users uncomfortable with these technologies. In this paper we attempt to build humanly interpretable visual models, allowing the user to visually understand the underlying semantic. We therefore propose a supervised multiple instance learning algorithm that selects as few as possible discriminant local features for a given object category. The method finds its roots in the lasso theory where a L1-regularization term is introduced in order to constraint the loss function, and subsequently produce sparser solutions. Efficient resolution of the lasso path is achieved through a boosting-like procedure inspired by BLasso algorithm. Quantitatively, our method achieves similar performance as current state-of-the-art, and qualitatively, it allows users to construct their own model from the original set of patches learned, thus allowing for more compound semantic queries.
Type de document :
Communication dans un congrès
ICMR'11 - First ACM International Conference on Multimedia Retrieval, Apr 2011, Trento, Italy. ACM, pp.21:1--21:8, 2011, 〈http://www.icmr2011.org/〉. 〈10.1145/1991996.1992017〉
Liste complète des métadonnées

Littérature citée [12 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00642232
Contributeur : Alexis Joly <>
Soumis le : jeudi 17 novembre 2011 - 15:58:34
Dernière modification le : jeudi 11 janvier 2018 - 16:20:54
Document(s) archivé(s) le : lundi 5 décembre 2016 - 09:14:02

Fichier

ICMR-rebai.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Ahmed Rebai, Alexis Joly, Nozha Boujemaa. Interpretable Visual Models for Human Perception-Based Object Retrieval. ICMR'11 - First ACM International Conference on Multimedia Retrieval, Apr 2011, Trento, Italy. ACM, pp.21:1--21:8, 2011, 〈http://www.icmr2011.org/〉. 〈10.1145/1991996.1992017〉. 〈hal-00642232〉

Partager

Métriques

Consultations de la notice

397

Téléchargements de fichiers

218