Person re-identification in multi-camera system by signature based on interest point descriptors collected on short video sequences

Abstract : We present and evaluate a person re-identification scheme for multi-camera surveillance system. Our approach uses matching of signatures based on interest-points descriptors collected on short video sequences. One of the originalities of our method is to accumulate interest points on several sufficiently time-spaced images during person tracking within each camera, in order to capture appearance variability. A first experimental evaluation conducted on a publicly available set of low-resolution videos in a commercial mall shows very promising inter-camera person re-identification performances (a precision of 82% for a recall of 78%). It should also be noted that our matching method is very fast: ~ 1/8s for re-identification of one target person among 10 previously seen persons, and a logarithmic dependence with the number of stored person models, making reidentification among hundreds of persons computationally feasible in less than ~ 1/5 second.
Type de document :
Communication dans un congrès
2nd ACM/IEEE International Conference on Distributed Smart Cameras (ICDSC-08), Sep 2008, Stanford, Palo Alto, United States. 2008
Liste complète des métadonnées

Littérature citée [13 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00332032
Contributeur : Fabien Moutarde <>
Soumis le : lundi 20 octobre 2008 - 11:54:59
Dernière modification le : mardi 27 mars 2018 - 16:06:18
Document(s) archivé(s) le : lundi 7 juin 2010 - 18:21:13

Fichier

MinesParisTech_PersonReidentif...
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00332032, version 1

Collections

Citation

Omar Hamdoun, Fabien Moutarde, Bogdan Stanciulescu, Bruno Steux. Person re-identification in multi-camera system by signature based on interest point descriptors collected on short video sequences. 2nd ACM/IEEE International Conference on Distributed Smart Cameras (ICDSC-08), Sep 2008, Stanford, Palo Alto, United States. 2008. 〈inria-00332032〉

Partager

Métriques

Consultations de la notice

418

Téléchargements de fichiers

1272