A contextual dissimilarity measure for accurate and efficient image search

Hervé Jégou 1 Harzallah Hedi 1 Cordelia Schmid 1
1 LEAR - Learning and recognition in vision
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : In this paper we present two contributions to improve accuracy and speed of an image search system based on bag-of-features: a contextual dissimilarity measure (CDM) and an efficient search structure for visual word vectors. Our measure (CDM) takes into account the local distribution of the vectors and iteratively estimates distance correcting terms. These terms are subsequently used to update an existing distance, thereby modifying the neighborhood structure. Experimental results on the Nister-Stewenius dataset show that our approach significantly outperforms the state-of-the-art in terms of accuracy. Our efficient search structure for visual word vectors is a two-level scheme using inverted files. The first level partitions the image set into clusters of images. At query time, only a subset of clusters of the second level has to be searched. This method allows fast querying in large sets of images. We valuate the gain in speed and the loss in accuracy on large datasets (up to 1 million images).
Type de document :
Communication dans un congrès
CVPR 2007 - Conference on Computer Vision & Pattern Recognition, Jun 2007, Minneapolis, United States. IEEE Computer society, pp.1-8, 2007, 〈http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4269995〉. 〈10.1109/CVPR.2007.382970〉
Liste complète des métadonnées


https://hal.inria.fr/inria-00394210
Contributeur : Hervé Jégou <>
Soumis le : mardi 15 mars 2011 - 14:38:06
Dernière modification le : mercredi 11 avril 2018 - 01:59:37
Document(s) archivé(s) le : jeudi 16 juin 2011 - 02:20:08

Identifiants

Collections

Citation

Hervé Jégou, Harzallah Hedi, Cordelia Schmid. A contextual dissimilarity measure for accurate and efficient image search. CVPR 2007 - Conference on Computer Vision & Pattern Recognition, Jun 2007, Minneapolis, United States. IEEE Computer society, pp.1-8, 2007, 〈http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4269995〉. 〈10.1109/CVPR.2007.382970〉. 〈inria-00394210〉

Partager

Métriques

Consultations de la notice

438

Téléchargements de fichiers

706