GANSpace: Discovering Interpretable GAN Controls
Erik Härkönen, Aaron Hertzmann, Jaakko Lehtinen, Sylvain Paris
Abstract
This paper describes a simple technique to analyze Generative Adversarial Networks (GANs) and create interpretable controls for image synthesis, such as change of viewpoint, aging, lighting, and time of day. We identify important latent directions based on Principal Component Analysis (PCA) applied either in latent space or feature space. Then, we show that a large number of interpretable controls can be defined by layer-wise perturbation along the principal directions. Moreover, we show that BigGAN can be controlled with layer-wise inputs in a StyleGAN-like manner. We show results on different GANs trained on various datasets, and demonstrate good qualitative matches to edit directions found through earlier supervised approaches.
BibTeX
@inproceedings{NEURIPS2020_6fe43269,
author = {H\"{a}rk\"{o}nen, Erik and Hertzmann, Aaron and Lehtinen, Jaakko and Paris, Sylvain},
booktitle = {Advances in Neural Information Processing Systems},
editor = {H. Larochelle and M. Ranzato and R. Hadsell and M.F. Balcan and H. Lin},
pages = {9841--9850},
publisher = {Curran Associates, Inc.},
title = {GANSpace: Discovering Interpretable GAN Controls},
url = {https://proceedings.neurips.cc/paper_files/paper/2020/file/6fe43269967adbb64ec6149852b5cc3e-Paper.pdf},
volume = {33},
year = {2020}
}