arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2305.13625cs.CVcs.CR
收藏

DiffProtect: 用扩散模型生成对抗示例以保护面部隐私

DiffProtect: Generate Adversarial Examples with Diffusion Models for Facial Privacy Protection

Johns Hopkins University(约翰霍普金斯大学);City University of Hong Kong(香港城市大学);University of St Andrews(圣安德鲁大学)

Jiang Liu, Chun Pong Lau, Zhongliang Guo, Yuxiang Guo, Zhaoyang Wang, Rama Chellappa

AI总结:

DiffProtect利用扩散模型生成高质量对抗示例,提升面部隐私保护的视觉质量和攻击成功率

摘要:

日益普及的面部识别(FR)系统引发了严重的个人隐私问题,尤其是对于数十亿在社交媒体上公开分享照片的用户。已有尝试利用对抗攻击生成加密面部图像来保护个人免受未经授权的FR系统识别。然而,现有方法存在视觉质量差或攻击成功率低的问题,限制了其实用性。最近,扩散模型在图像生成中取得了巨大成功。在本工作中,我们提出一个问题:扩散模型能否用于生成对抗示例以提高视觉质量和攻击性能?我们提出了DiffProtect,利用扩散自编码器在FR系统上生成具有语义意义的扰动。大量实验表明,DiffProtect生成的加密图像比最先进的方法更自然,同时实现了显著更高的攻击成功率,例如在CelebA-HQ和FFHQ数据集上分别达到24.5%和25.1%的绝对提升。

英文摘要:

The increasingly pervasive facial recognition (FR) systems raise serious concerns about personal privacy, especially for billions of users who have publicly shared their photos on social media. Several attempts have been made to protect individuals from being identified by unauthorized FR systems utilizing adversarial attacks to generate encrypted face images. However, existing methods suffer from poor visual quality or low attack success rates, which limit their utility. Recently, diffusion models have achieved tremendous success in image generation. In this work, we ask: can diffusion models be used to generate adversarial examples to improve both visual quality and attack performance? We propose DiffProtect, which utilizes a diffusion autoencoder to generate semantically meaningful perturbations on FR systems. Extensive experiments demonstrate that DiffProtect produces more natural-looking encrypted images than state-of-the-art methods while achieving significantly higher attack success rates, e.g., 24.5% and 25.1% absolute improvements on the CelebA-HQ and FFHQ datasets.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑