Does CLIP Know My Face?
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2022
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909247616319488 |
|---|---|
| author | Hintersdorf, Dominik Struppek, Lukas Brack, Manuel Friedrich, Felix Schramowski, Patrick Kersting, Kristian |
| author_facet | Hintersdorf, Dominik Struppek, Lukas Brack, Manuel Friedrich, Felix Schramowski, Patrick Kersting, Kristian |
| contents | With the rise of deep learning in various applications, privacy concerns around the protection of training data have become a critical area of research. Whereas prior studies have focused on privacy risks in single-modal models, we introduce a novel method to assess privacy for multi-modal models, specifically vision-language models like CLIP. The proposed Identity Inference Attack (IDIA) reveals whether an individual was included in the training data by querying the model with images of the same person. Letting the model choose from a wide variety of possible text labels, the model reveals whether it recognizes the person and, therefore, was used for training. Our large-scale experiments on CLIP demonstrate that individuals used for training can be identified with very high accuracy. We confirm that the model has learned to associate names with depicted individuals, implying the existence of sensitive information that can be extracted by adversaries. Our results highlight the need for stronger privacy protection in large-scale models and suggest that IDIAs can be used to prove the unauthorized use of data for training and to enforce privacy laws. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2209_07341 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Does CLIP Know My Face? Hintersdorf, Dominik Struppek, Lukas Brack, Manuel Friedrich, Felix Schramowski, Patrick Kersting, Kristian Machine Learning Cryptography and Security Computer Vision and Pattern Recognition With the rise of deep learning in various applications, privacy concerns around the protection of training data have become a critical area of research. Whereas prior studies have focused on privacy risks in single-modal models, we introduce a novel method to assess privacy for multi-modal models, specifically vision-language models like CLIP. The proposed Identity Inference Attack (IDIA) reveals whether an individual was included in the training data by querying the model with images of the same person. Letting the model choose from a wide variety of possible text labels, the model reveals whether it recognizes the person and, therefore, was used for training. Our large-scale experiments on CLIP demonstrate that individuals used for training can be identified with very high accuracy. We confirm that the model has learned to associate names with depicted individuals, implying the existence of sensitive information that can be extracted by adversaries. Our results highlight the need for stronger privacy protection in large-scale models and suggest that IDIAs can be used to prove the unauthorized use of data for training and to enforce privacy laws. |
| title | Does CLIP Know My Face? |
| topic | Machine Learning Cryptography and Security Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2209.07341 |