Unpaired Image-to-Image Translation with Limited Data to Reveal Subtle Phenotypes - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Unpaired Image-to-Image Translation with Limited Data to Reveal Subtle Phenotypes

Kevin Daupin
Aurélie de Thonel
  • Fonction : Auteur
  • PersonId : 1035993
Valerie Mezger-Lallemand
  • Fonction : Auteur
Auguste Genovesio

Résumé

Unpaired image-to-image translation methods aim at learning a mapping of images from a source domain to a target domain. Recently, these methods showed to be very useful in biological applications to display subtle phenotypic cell variations otherwise invisible to the human eye. However, while most microscopy experiments remain limited in the number of images they can produce, current models require a large number of images to be trained. In this work, we present an improved CycleGAN architecture that employs self-supervised discriminators to alleviate the need for numerous images. We demonstrate quantitatively and qualitatively that the proposed approach outperforms the CycleGAN baseline, including when it is combined with differentiable augmentations. We also provide results obtained with small biological datasets on obvious and non-obvious cell phenotype variations demonstrating a straightforward application of this method.
Fichier principal
Vignette du fichier
2022_Bourou_NeurIPS_SSL.pdf (1.64 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04081464 , version 1 (25-04-2023)

Identifiants

  • HAL Id : hal-04081464 , version 1

Citer

Anis Bourou, Kevin Daupin, Veronique Dubreuil, Aurélie de Thonel, Valerie Mezger-Lallemand, et al.. Unpaired Image-to-Image Translation with Limited Data to Reveal Subtle Phenotypes. Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS 202), Nov 2022, La Nouvelle Orléans, LA, United States. ⟨hal-04081464⟩
15 Consultations
2 Téléchargements

Partager

Gmail Facebook X LinkedIn More