Anti-Reference: Universal and Immediate Defense Against Reference-Based Generation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866929619991527424 |
|---|---|
| author | Song, Yiren Lou, Shengtao Liu, Xiaokang Ci, Hai Yang, Pei Liu, Jiaming Shou, Mike Zheng |
| author_facet | Song, Yiren Lou, Shengtao Liu, Xiaokang Ci, Hai Yang, Pei Liu, Jiaming Shou, Mike Zheng |
| contents | Diffusion models have revolutionized generative modeling with their exceptional ability to produce high-fidelity images. However, misuse of such potent tools can lead to the creation of fake news or disturbing content targeting individuals, resulting in significant social harm. In this paper, we introduce Anti-Reference, a novel method that protects images from the threats posed by reference-based generation techniques by adding imperceptible adversarial noise to the images. We propose a unified loss function that enables joint attacks on fine-tuning-based customization methods, non-fine-tuning customization methods, and human-centric driving methods. Based on this loss, we train a Adversarial Noise Encoder to predict the noise or directly optimize the noise using the PGD method. Our method shows certain transfer attack capabilities, effectively challenging both gray-box models and some commercial APIs. Extensive experiments validate the performance of Anti-Reference, establishing a new benchmark in image security. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_05980 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Anti-Reference: Universal and Immediate Defense Against Reference-Based Generation Song, Yiren Lou, Shengtao Liu, Xiaokang Ci, Hai Yang, Pei Liu, Jiaming Shou, Mike Zheng Computer Vision and Pattern Recognition Diffusion models have revolutionized generative modeling with their exceptional ability to produce high-fidelity images. However, misuse of such potent tools can lead to the creation of fake news or disturbing content targeting individuals, resulting in significant social harm. In this paper, we introduce Anti-Reference, a novel method that protects images from the threats posed by reference-based generation techniques by adding imperceptible adversarial noise to the images. We propose a unified loss function that enables joint attacks on fine-tuning-based customization methods, non-fine-tuning customization methods, and human-centric driving methods. Based on this loss, we train a Adversarial Noise Encoder to predict the noise or directly optimize the noise using the PGD method. Our method shows certain transfer attack capabilities, effectively challenging both gray-box models and some commercial APIs. Extensive experiments validate the performance of Anti-Reference, establishing a new benchmark in image security. |
| title | Anti-Reference: Universal and Immediate Defense Against Reference-Based Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.05980 |