EvolGAN: Evolutionary Generative Adversarial Networks
This method addresses image quality issues in GANs for specific domains like art and fashion, but it is incremental as it builds on existing GAN frameworks.
The paper tackled the problem of generating high-quality images from GANs trained on small or difficult datasets by using evolutionary methods to search the latent space, resulting in significant improvements in image quality with human preference rates up to 83.7% across various datasets.
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.