Learning finer-class networks for universal representations - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

Learning finer-class networks for universal representations

Résumé

Many real-world visual recognition use-cases can not directly benefit from state-ofthe-art CNN-based approaches because of the lack of many annotated data. The usual approach to deal with this is to transfer a representation pre-learned on a large annotated source-task onto a target-task of interest. This raises the question of how well the original representation is "universal", that is to say directly adapted to many different target-tasks. To improve such universality, the state-of-the-art consists in training networks on a diversified source problem, that is modified either by adding generic or specific categories to the initial set of categories. In this vein, we proposed a method that exploits finer-classes than the most specific ones existing, for which no annotation is available. We rely on unsupervised learning and a bottom-up split and merge strategy. We show that our method learns more universal representations than state-of-the-art, leading to significantly better results on 10 target-tasks from multiple domains, using several network architectures, either alone or combined with networks learned at a coarser semantic level.
Fichier principal
Vignette du fichier
1021.pdf (1.33 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Licence : CC BY - Paternité

Dates et versions

hal-04315371 , version 1 (30-11-2023)

Licence

Paternité - Pas d'utilisation commerciale - Pas de modification

Identifiants

Citer

Julien Girard, Youssef Tamaazousti, Hervé Le Borgne, Céline Hudelot. Learning finer-class networks for universal representations. BMVC 2018 - the 29th British Machine Vision Conference, Sep 2018, Newcastle, United Kingdom. ⟨10.48550/arXiv.1810.02126⟩. ⟨hal-04315371⟩
16 Consultations
5 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More