Adv-Diffusion: Imperceptible Adversarial Face Identity Attack via Latent Diffusion Model
This addresses security vulnerabilities in face recognition systems by improving attack stealth and transferability, representing an incremental advance over existing adversarial generation methods.
The paper tackles the problem of generating imperceptible adversarial attacks on face recognition models by proposing Adv-Diffusion, a framework that uses a latent diffusion model to create realistic adversarial images, achieving superior performance with high transferability and stealthiness compared to state-of-the-art methods on FFHQ and CelebA-HQ datasets.
Adversarial attacks involve adding perturbations to the source image to cause misclassification by the target model, which demonstrates the potential of attacking face recognition models. Existing adversarial face image generation methods still can't achieve satisfactory performance because of low transferability and high detectability. In this paper, we propose a unified framework Adv-Diffusion that can generate imperceptible adversarial identity perturbations in the latent space but not the raw pixel space, which utilizes strong inpainting capabilities of the latent diffusion model to generate realistic adversarial images. Specifically, we propose the identity-sensitive conditioned diffusion generative model to generate semantic perturbations in the surroundings. The designed adaptive strength-based adversarial perturbation algorithm can ensure both attack transferability and stealthiness. Extensive qualitative and quantitative experiments on the public FFHQ and CelebA-HQ datasets prove the proposed method achieves superior performance compared with the state-of-the-art methods without an extra generative model training process. The source code is available at https://github.com/kopper-xdu/Adv-Diffusion.