EvolGAN: Evolutionary Generative Adversarial Networks

by   Baptiste Roziere, et al.

We propose to use a quality estimator and evolutionary methods to search the latent space of generative adversarial networks trained on small, difficult datasets, or both. The new method leads to the generation of significantly higher quality images while preserving the original generator's diversity. Human raters preferred an image from the new version with frequency 83.7pc for Cats, 74pc for FashionGen, 70.4pc for Horses, and 69.2pc for Artworks, and minor improvements for the already excellent GANs for faces. This approach applies to any quality scorer and GAN generator.



page 2

page 9

page 11


IVE-GAN: Invariant Encoding Generative Adversarial Networks

Generative adversarial networks (GANs) are a powerful framework for gene...

A Style-Based Generator Architecture for Generative Adversarial Networks

We propose an alternative generator architecture for generative adversar...

Reparameterized Sampling for Generative Adversarial Networks

Recently, sampling methods have been successfully applied to enhance the...

Evolution of Images with Diversity and Constraints Using a Generator Network

Evolutionary search has been extensively used to generate artistic image...

Generative Adversarial Networks for Non-Raytraced Global Illumination on Older GPU Hardware

We give an overview of the different rendering methods and we demonstrat...

Evolutionary latent space search for driving human portrait generation

This article presents an evolutionary approach for synthetic human portr...

SWAGAN: A Style-based Wavelet-driven Generative Model

In recent years, considerable progress has been made in the visual quali...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.