Segmentation in Style: Unsupervised Semantic Image Segmentation with Stylegan and CLIP
This addresses the problem of automating semantic segmentation without human labels, which is incremental as it builds on existing generative and vision-language models.
The paper tackles unsupervised semantic image segmentation by using pretrained StyleGAN2 and CLIP to discover semantic classes from generated images, then training a segmentation model on synthetic data that generalizes to real images, achieving state-of-the-art results on public datasets.
We introduce a method that allows to automatically segment images into semantically meaningful regions without human supervision. Derived regions are consistent across different images and coincide with human-defined semantic classes on some datasets. In cases where semantic regions might be hard for human to define and consistently label, our method is still able to find meaningful and consistent semantic classes. In our work, we use pretrained StyleGAN2 generative model: clustering in the feature space of the generative model allows to discover semantic classes. Once classes are discovered, a synthetic dataset with generated images and corresponding segmentation masks can be created. After that a segmentation model is trained on the synthetic dataset and is able to generalize to real images. Additionally, by using CLIP we are able to use prompts defined in a natural language to discover some desired semantic classes. We test our method on publicly available datasets and show state-of-the-art results.