Finding an unsupervised image segmenter in each of your deep generative models

Conference item : Poster

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.

Authors: Melas-Kyriazi, L; Rupprecht, C; Laina, I; Vedaldi, A

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: OpenReview
• Host title: International Conference on Learning Representations
• Article number: 1643
• Publication date: 2021-09-29
• Acceptance date: 2022-01-24
• Event title: Tenth International Conference on Learning Representations (ICLR 2022)
• Event location: Virtual event
• Event website: https://iclr.cc/Conferences/2022/
• Event start date: 2022-04-25
• Event end date: 2022-04-29
• Language: English
• Keywords: object segmentation; deep learning; unsupervised; FFR; segmentation; generative
• Subtype: Poster