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 2401.01575cs.CV

通过梯度累积增强不可见面部隐私保护罩的泛化能力

Enhancing Generalization of Invisible Facial Privacy Cloak via Gradient Accumulation

Inspur (Beijing) Electronic Information Industry Co., Ltd(浪潮(北京)电子信息产业有限公司)

Xuannan Liu, Yaoyao Zhong, Weihong Deng, Hongzhi Shi, Xingchen Cui, Yunfeng Yin, Dongchao Wen

AI总结:

针对人脸识别系统引发的隐私担忧,本文指出现有类通用对抗性隐私保护罩在大/小批量优化中存在的困境,提出利用梯度累积(GA)聚合小批量梯度以增强稳定性并减少量化操作,在Privacy-Commons数据集上有效抵御了黑盒人脸识别模型。

摘要:

社交媒体和人脸识别(FR)系统的繁荣增加了人们对隐私和安全的担忧。一种新型对抗性隐私保护罩(类通用型)可应用于普通用户的所有图像,以防止恶意 FR 系统获取其身份信息。在本研究中,我们发现了现有方法中的优化困境——大批量优化中的局部最优问题和小批量优化中的梯度信息消除问题。为解决这些问题,我们提出梯度累积(GA),将多个小批量梯度聚合为单步迭代梯度,以增强梯度稳定性并减少量化操作的使用。实验表明,我们提出的方法在 Privacy-Commons 数据集上针对黑盒人脸识别模型取得了高性能。

英文摘要:

The blooming of social media and face recognition (FR) systems has increased people's concern about privacy and security. A new type of adversarial privacy cloak (class-universal) can be applied to all the images of regular users, to prevent malicious FR systems from acquiring their identity information. In this work, we discover the optimization dilemma in the existing methods -- the local optima problem in large-batch optimization and the gradient information elimination problem in small-batch optimization. To solve these problems, we propose Gradient Accumulation (GA) to aggregate multiple small-batch gradients into a one-step iterative gradient to enhance the gradient stability and reduce the usage of quantization operations. Experiments show that our proposed method achieves high performance on the Privacy-Commons dataset against black-box face recognition models.

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。

↑