Abstract

Recent research has shown that numerous human-interpretable directions exist in the latent space of GANs. In this paper, we develop an automatic procedure for finding directions that lead to foreground-background image separation, and we use these directions to train an image segmentation model without human supervision. Our method is generator-agnostic, producing strong segmentation results with a wide range of different GAN architectures. Furthermore, by leveraging GANs pretrained on large datasets such as ImageNet, we are able to segment images from a range of domains without further training or finetuning. Evaluating our method on image segmentation benchmarks, we compare favorably to prior work while using neither human supervision nor access to the training data. Broadly, our results demonstrate that automatically extracting foreground-background structure from pretrained deep generative models can serve as a remarkably effective substitute for human supervision.

Examples

Examples of interpolations in GAN latent space

Examples of segmentations produced by the final segmentation model, which was trained entirely on GAN-generated images

Citation

Acknowledgements

C. R. is supported by Innovate UK (project 71653) on behalf of UK Research and Innovation (UKRI) and by the European Research Council (ERC) IDIU-638009. I. L. is supported by the EPSRC programme grant Seebibyte EP/M013774/1 and ERC starting grant IDIU-638009. A. V. is funded by ERC grant IDIU-638009.