Decomposing Bag of Words Histograms

Ankit Gandhi 1 Karteek Alahari 2, 3, 4 C.V. Jawahar 1
2 WILLOW - Models of visual object recognition and scene understanding
DI-ENS - Département d'informatique de l'École normale supérieure, Inria Paris-Rocquencourt, CNRS - Centre National de la Recherche Scientifique : UMR8548
4 LEAR - Learning and recognition in vision
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : We aim to decompose a global histogram representation of an image into histograms of its associated objects and regions. This task is formulated as an optimization problem, given a set of linear classifiers, which can effectively discriminate the object categories present in the image. Our decomposition bypasses harder problems associated with accurately localizing and segmenting objects. We evaluate our method on a wide variety of composite histograms, and also compare it with MRF-based solutions. In addition to merely measuring the accuracy of decomposition, we also show the utility of the estimated object and background histograms for the task of image classification on the PASCAL VOC 2007 dataset.
Type de document :
Communication dans un congrès
ICCV - IEEE International Conference on Computer Vision, Dec 2013, Sydney, Australia. IEEE, pp.305-312, 2013, 〈10.1109/ICCV.2013.45〉
Liste complète des métadonnées

Littérature citée [35 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00874895
Contributeur : Karteek Alahari <>
Soumis le : vendredi 18 octobre 2013 - 18:11:48
Dernière modification le : lundi 28 mai 2018 - 15:10:02
Document(s) archivé(s) le : dimanche 19 janvier 2014 - 04:31:01

Fichier

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

Identifiants

Collections

Citation

Ankit Gandhi, Karteek Alahari, C.V. Jawahar. Decomposing Bag of Words Histograms. ICCV - IEEE International Conference on Computer Vision, Dec 2013, Sydney, Australia. IEEE, pp.305-312, 2013, 〈10.1109/ICCV.2013.45〉. 〈hal-00874895〉

Partager

Métriques

Consultations de la notice

1663

Téléchargements de fichiers

913