Domain-Unified Prompt Representations for Source-Free Domain Generalization
This work addresses the problem of domain generalization for AI models in open-world scenarios, such as science fiction and pixelate styles, but it is incremental as it builds on existing vision-language models.
The paper tackles the challenge of source-free domain generalization (SFDG) by proposing a method that uses large-scale vision-language models like CLIP to generate diverse prompts from a domain bank, enabling domain-unified representations for open-world scenarios. It achieves competitive performance on mainstream DG datasets such as PACS, VLCS, OfficeHome, and DomainNet, with results comparable to state-of-the-art methods that require source domain data.
Domain generalization (DG), aiming to make models work on unseen domains, is a surefire way toward general artificial intelligence. Limited by the scale and diversity of current DG datasets, it is difficult for existing methods to scale to diverse domains in open-world scenarios (e.g., science fiction and pixelate style). Therefore, the source-free domain generalization (SFDG) task is necessary and challenging. To address this issue, we propose an approach based on large-scale vision-language pretraining models (e.g., CLIP), which exploits the extensive domain information embedded in it. The proposed scheme generates diverse prompts from a domain bank that contains many more diverse domains than existing DG datasets. Furthermore, our method yields domain-unified representations from these prompts, thus being able to cope with samples from open-world domains. Extensive experiments on mainstream DG datasets, namely PACS, VLCS, OfficeHome, and DomainNet, show that the proposed method achieves competitive performance compared to state-of-the-art (SOTA) DG methods that require source domain data for training. Besides, we collect a small datasets consists of two domains to evaluate the open-world domain generalization ability of the proposed method. The source code and the dataset will be made publicly available at https://github.com/muse1998/Source-Free-Domain-Generalization