Small objects query suggestion in a large web-image collection

Pierre Letessier 1, 2 Nicolas Hervé 1 Champ Julien 2 Alexis Joly 2 Olivier Buisson 1 Amel Hamzaoui 2
2 ZENITH - Scientific Data Management
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier, CRISAM - Inria Sophia Antipolis - Méditerranée
Abstract : State-of-the-art visual search methods allow retrieving efficiently small rigid objects in very large image datasets (e.g. logos, paintings, etc.). User's perception of the classical query-by-window paradigm is however affected by the fact that many submitted queries actually return nothing or only junk results. We demonstrate in this demo that the perception can be radically different if the objects of interest are rather suggested to the user by pre-computing relevant clusters of instances. Impressive results involving very small objects discovered in a web collection of 110K images are demonstrated through a simple interactive GUI.
Type de document :
Communication dans un congrès
MM'13: ACM Multimedia, Oct 2013, Barcelone, Spain. ACM, 2013, 〈10.1145/2502081.2502248〉
Liste complète des métadonnées

https://hal.inria.fr/hal-00908891
Contributeur : Alexis Joly <>
Soumis le : lundi 25 novembre 2013 - 14:40:20
Dernière modification le : vendredi 12 janvier 2018 - 01:53:28

Identifiants

Citation

Pierre Letessier, Nicolas Hervé, Champ Julien, Alexis Joly, Olivier Buisson, et al.. Small objects query suggestion in a large web-image collection. MM'13: ACM Multimedia, Oct 2013, Barcelone, Spain. ACM, 2013, 〈10.1145/2502081.2502248〉. 〈hal-00908891〉

Partager

Métriques

Consultations de la notice

195