Local Convolutional Features with Unsupervised Training for Image Retrieval

Mattis Paulin 1 Matthijs Douze 1 Zaid Harchaoui 1 Julien Mairal 1 Florent Perronnin 2 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 : Patch-level descriptors underlie several important computer vision tasks, such as stereo-matching or content-based image retrieval. We introduce a deep convolutional architecture that yields patch-level descriptors, as an alternative to the popular SIFT descriptor for image retrieval. The proposed family of descriptors, called Patch-CKN, adapt the recently introduced Convolutional Kernel Network (CKN), an unsupervised framework to learn convolutional architectures. We present a comparison framework to benchmark current deep convolutional approaches along with Patch-CKN for both patch and image retrieval, including our novel ``RomePatches'' dataset. Patch-CKN descriptors yield competitive results compared to supervised CNNs alternatives on patch and image retrieval.
Type de document :
Communication dans un congrès
ICCV - IEEE International Conference on Computer Vision, Dec 2015, Santiago, Chile. IEEE, pp.91-99, 〈10.1109/ICCV.2015.19〉
Liste complète des métadonnées

Littérature citée [42 références]  Voir  Masquer  Télécharger


https://hal.inria.fr/hal-01207966
Contributeur : Thoth Team <>
Soumis le : jeudi 1 octobre 2015 - 16:27:12
Dernière modification le : vendredi 11 août 2017 - 11:52:22
Document(s) archivé(s) le : samedi 2 janvier 2016 - 11:20:59

Fichiers

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

Identifiants

Collections

Citation

Mattis Paulin, Matthijs Douze, Zaid Harchaoui, Julien Mairal, Florent Perronnin, et al.. Local Convolutional Features with Unsupervised Training for Image Retrieval. ICCV - IEEE International Conference on Computer Vision, Dec 2015, Santiago, Chile. IEEE, pp.91-99, 〈10.1109/ICCV.2015.19〉. 〈hal-01207966〉

Partager

Métriques

Consultations de
la notice

1310

Téléchargements du document

10478