Improving Deep Metric Learning With Virtual Classes And Examples Mining - ETIS, équipe de recherche CELL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Improving Deep Metric Learning With Virtual Classes And Examples Mining

Résumé

In deep metric learning, the training procedure relies on sampling informative tuples. However, as the training procedure progresses, it becomes nearly impossible to sample relevant hard negative examples without proper mining strategies or generation-based methods. Recent work on hard negative generation have shown great promises to solve the mining problem. However, this generation process is difficult to tune and often leads to incorrectly labeled examples. To tackle this issue, we introduce MIRAGE, a generation-based method that relies on virtual classes entirely composed of enerated examples that act as buffer areas between the training classes. We empirically show that virtual classes significantly improve the results on popular datasets (Cub-200-2011 and Cars-196) compared to other generation methods
Fichier principal
Vignette du fichier
icip22_Mirage.pdf (882.55 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03740481 , version 1 (29-09-2022)

Identifiants

  • HAL Id : hal-03740481 , version 1

Citer

Pierre Jacob, David Picard, Aymeric Histace. Improving Deep Metric Learning With Virtual Classes And Examples Mining. International Conference on Image Processing (IEEE ICIP), IEEE, Oct 2022, Bordeaux, France. ⟨hal-03740481⟩
88 Consultations
37 Téléchargements

Partager

Gmail Facebook X LinkedIn More