Mining on Manifolds: Metric Learning without Labels

Abstract : In this work we present a novel unsupervised framework for hard training example mining. The only input to the method is a collection of images relevant to the target application and a meaningful initial representation, provided e.g. by pre-trained CNN. Positive examples are distant points on a single manifold, while negative examples are nearby points on different manifolds. Both types of examples are revealed by disagreements between Euclidean and manifold similarities. The discovered examples can be used in training with any discriminative loss. The method is applied to unsupervised fine-tuning of pre-trained networks for fine-grained classification and particular object retrieval. Our models are on par or are outperforming prior models that are fully or partially supervised.
Type de document :
Communication dans un congrès
IEEE Computer Vision and Pattern Recognition Conference, Jun 2018, Salt Lake City, United States
Liste complète des métadonnées

https://hal.inria.fr/hal-01843085
Contributeur : Yannis Avrithis <>
Soumis le : mercredi 18 juillet 2018 - 14:46:47
Dernière modification le : vendredi 20 juillet 2018 - 01:11:32

Fichier

C108.cvpr18.mom.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01843085, version 1

Citation

Ahmet Iscen, Giorgos Tolias, Yannis Avrithis, Ondřej Chum. Mining on Manifolds: Metric Learning without Labels. IEEE Computer Vision and Pattern Recognition Conference, Jun 2018, Salt Lake City, United States. 〈hal-01843085〉

Partager

Métriques

Consultations de la notice

109

Téléchargements de fichiers

13