How good is my GAN?

Konstantin Shmelkov 1 Cordelia Schmid 1 Karteek Alahari 1
1 Thoth - Apprentissage de modèles à partir de données massives
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann
Abstract : Generative adversarial networks (GANs) are one of the most popular methods for generating images today. While impressive results have been validated by visual inspection, a number of quantitative criteria have emerged only recently. We argue here that the existing ones are insufficient and need to be in adequation with the task at hand. In this paper we introduce two measures based on image classification---GAN-train and GAN-test, which approximate the recall (diversity) and precision (quality of the image) of GANs respectively. We evaluate a number of recent GAN approaches based on these two measures and demonstrate a clear difference in performance. Furthermore, we observe that the increasing difficulty of the dataset, from CIFAR10 over CIFAR100 to ImageNet, shows an inverse correlation with the quality of the GANs, as clearly evident from our measures.
Liste complète des métadonnées

https://hal.inria.fr/hal-01850447
Contributor : Karteek Alahari <>
Submitted on : Friday, July 27, 2018 - 1:15:33 PM
Last modification on : Thursday, February 7, 2019 - 4:21:59 PM
Document(s) archivé(s) le : Sunday, October 28, 2018 - 1:24:16 PM

File

ganeval.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Konstantin Shmelkov, Cordelia Schmid, Karteek Alahari. How good is my GAN?. ECCV 2018 - European Conference on Computer Vision, Sep 2018, Munich, Germany. pp.218-234, ⟨10.1007/978-3-030-01216-8_14⟩. ⟨hal-01850447⟩

Share

Metrics

Record views

1223

Files downloads

1388