Fake it till you make it: Learning transferable representations from synthetic ImageNet clones - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Fake it till you make it: Learning transferable representations from synthetic ImageNet clones

Résumé

Recent image generation models such as Stable Diffusion have exhibited an impressive ability to generate fairly realistic images starting from a simple text prompt. Could such models render real images obsolete for training image prediction models? In this paper, we answer part of this provocative question by investigating the need for real images when training models for ImageNet classification. Provided only with the class names that have been used to build the dataset, we explore the ability of Stable Diffusion to generate synthetic clones of ImageNet and measure how useful these are for training classification models from scratch. We show that with minimal and class-agnostic prompt engineering, ImageNet clones are able to close a large part of the gap between models produced by synthetic images and models trained with real images, for the several standard classification benchmarks that we consider in this study. More importantly, we show that models trained on synthetic images exhibit strong generalization properties and perform on par with models trained on real data for transfer. Project page: https://europe.naverlabs.com/imagenet-sd
Fichier principal
Vignette du fichier
sariyildiz23a.pdf (3.92 Mo) Télécharger le fichier
sariyildiz23a-supp.pdf (13.97 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03916262 , version 1 (30-12-2022)
hal-03916262 , version 2 (03-04-2023)

Identifiants

Citer

Mert Bulent Sariyildiz, Karteek Alahari, Diane Larlus, Yannis Kalantidis. Fake it till you make it: Learning transferable representations from synthetic ImageNet clones. CVPR 2023 - IEEE/CVF Conference on Computer Vision and Pattern Recognition, Jun 2023, Vancouver, Canada. pp.8011-8021, ⟨10.1109/CVPR52729.2023.00774⟩. ⟨hal-03916262v2⟩
757 Consultations
277 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More