GANSpace: Discovering Interpretable GAN Controls
This provides a simple, unsupervised method for interpretable image editing in GANs, which is incremental but useful for researchers and practitioners in generative modeling.
The paper tackles the problem of controlling image synthesis in GANs by discovering interpretable latent directions, such as viewpoint and lighting changes, using PCA-based techniques, and demonstrates that these controls match supervised methods across various GANs and datasets.
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 Components 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.