EvolGAN: Evolutionary Generative Adversarial Networks

Baptiste Roziere, Fabien Teytaud, Vlad Hosu, Hanhe Lin, Jeremy Rapin, Mariia Zameshina, Olivier Teytaud; Proceedings of the Asian Conference on Computer Vision (ACCV), 2020

Abstract

We propose to use a quality estimator and evolutionarymethods to search the latent space of generative adversarial networkstrained on small, difficult datasets, or both. The new method leads tothe generation of significantly higher quality images while preserving theoriginal generator's diversity. Human raters preferred an image from thenew version with frequency 83.7% for Cats, 74% for FashionGen, 70.4%for Horses, and 69.2% for Artworks - minor improvements for the alreadyexcellent GANs for faces. This approach applies to any quality scorer andGAN generator.

Related Material

[pdf] [arXiv]
[bibtex]
@InProceedings{Roziere_2020_ACCV, author = {Roziere, Baptiste and Teytaud, Fabien and Hosu, Vlad and Lin, Hanhe and Rapin, Jeremy and Zameshina, Mariia and Teytaud, Olivier}, title = {EvolGAN: Evolutionary Generative Adversarial Networks}, booktitle = {Proceedings of the Asian Conference on Computer Vision (ACCV)}, month = {November}, year = {2020} }