Abstract
The rise of generative image models leads to privacy concerns when it comes to the huge datasets used to train such models. This paper investigates the possibility of inferring if a set of face images was used for fine-tuning a Latent Diffusion Model (LDM). A Membership Inference Attack (MIA) method is presented for this task. Using generated auxiliary data for the training of the attack model leads to significantly better performance, and so does the use of watermarks. The guidance scale used for inference was found to have a significant influence. If a LDM is fine-tuned for long enough, the text prompt used for inference has no significant influence. The proposed MIA is found to be viable in a realistic black-box setup against LDMs fine-tuned on face-images.
Original language | English |
---|---|
Title of host publication | Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications |
Volume | 2 |
Publisher | SCITEPRESS Digital Library |
Publication date | 2025 |
Pages | 439-446 |
ISBN (Electronic) | 978-989-758-728-3 |
DOIs | |
Publication status | Published - 2025 |
Event | 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Porto, Portugal Duration: 26 Feb 2025 → 28 Feb 2025 |
Conference
Conference | 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications |
---|---|
Country/Territory | Portugal |
City | Porto |
Period | 26/02/2025 → 28/02/2025 |
Keywords
- Latent Diffusion Model
- Membership Inference Attack