Random Maximum Margin Hashing

Alexis Joly 1, 2, * Olivier Buisson 3
* Auteur correspondant
2 ZENITH - Scientific Data Management
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier, CRISAM - Inria Sophia Antipolis - Méditerranée
3 INA-SUP
INA - Institut National de l'Audiovisuel
Abstract : Following the success of hashing methods for multidimensional indexing, more and more works are interested in embedding visual feature space in compact hash codes. Such approaches are not an alternative to using index structures but a complementary way to reduce both the memory usage and the distance computation cost. Several data dependent hash functions have notably been proposed to closely fit data distribution and provide better selectivity than usual random projections such as LSH. However, improvements occur only for relatively small hash code sizes up to 64 or 128 bits. As discussed in the paper, this is mainly due to the lack of independence between the produced hash functions. We introduce a new hash function family that attempts to solve this issue in any kernel space. Rather than boosting the collision probability of close points, our method focus on data scattering. By training purely random splits of the data, regardless the closeness of the training samples, it is indeed possible to generate consistently more independent hash functions. On the other side, the use of large margin classifiers allows to maintain good generalization performances. Experiments show that our new Random Maximum Margin Hashing scheme (RMMH) outperforms four state-of-the-art hashing methods, notably in kernel spaces.
Type de document :
Communication dans un congrès
CVPR'11 - IEEE Computer Vision and Pattern Recognition, Jun 2011, Colorado springs, United States. IEEE, pp.873-880, 2011, 〈http://cvpr2011.org/〉. 〈10.1109/CVPR.2011.5995709〉
Liste complète des métadonnées

Littérature citée [23 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00642178
Contributeur : Alexis Joly <>
Soumis le : jeudi 17 novembre 2011 - 15:07:18
Dernière modification le : vendredi 25 mai 2018 - 12:02:03
Document(s) archivé(s) le : dimanche 4 décembre 2016 - 18:26:10

Fichier

rmmh-cvpr-certified-IEEE-eXpre...
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Alexis Joly, Olivier Buisson. Random Maximum Margin Hashing. CVPR'11 - IEEE Computer Vision and Pattern Recognition, Jun 2011, Colorado springs, United States. IEEE, pp.873-880, 2011, 〈http://cvpr2011.org/〉. 〈10.1109/CVPR.2011.5995709〉. 〈hal-00642178〉

Partager

Métriques

Consultations de la notice

519

Téléchargements de fichiers

1084