Panorama to panorama matching for location recognition

Abstract : Location recognition is commonly treated as visual instance retrieval on "street view" imagery. The dataset items and queries are panoramic views, i.e.groups of images taken at a single location. This work introduces a novel panorama-to-panorama matching process, either by aggregating features of individual images in a group or by explicitly constructing a larger panorama. In either case, multiple views are used as queries. We reach near perfect location recognition on a standard benchmark with only four query views.
Type de document :
Communication dans un congrès
ACM International Conference on Multimedia Retrieval (ICMR) 2017, Jun 2017, Bucharest, Romania. <10.1145/3078971.3079033>
Liste complète des métadonnées


https://hal.inria.fr/hal-01511793
Contributeur : Ahmet Iscen <>
Soumis le : vendredi 21 avril 2017 - 15:56:45
Dernière modification le : mercredi 2 août 2017 - 10:10:38
Document(s) archivé(s) le : samedi 22 juillet 2017 - 13:54:53

Fichier

Iscen_icmr17.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Ahmet Iscen, Giorgos Tolias, Yannis Avrithis, Teddy Furon, Ondřej Chum. Panorama to panorama matching for location recognition. ACM International Conference on Multimedia Retrieval (ICMR) 2017, Jun 2017, Bucharest, Romania. <10.1145/3078971.3079033>. <hal-01511793>

Partager

Métriques

Consultations de
la notice

364

Téléchargements du document

90