AdvCloak: Customized Adversarial Cloak for Privacy Protection

TL;DR

AdvCloak is a novel framework that creates personalized adversarial masks to protect facial privacy, balancing natural appearance and robustness against facial variations through a two-stage training process.

Contribution

It introduces a two-stage training strategy for generative adversarial networks to produce customizable, natural-looking privacy masks that generalize across diverse facial features.

Findings

Outperforms existing methods in efficiency and effectiveness

Maintains high image naturalness while enhancing privacy protection

Effective on both common and celebrity datasets

Abstract

With extensive face images being shared on social media, there has been a notable escalation in privacy concerns. In this paper, we propose AdvCloak, an innovative framework for privacy protection using generative models. AdvCloak is designed to automatically customize class-wise adversarial masks that can maintain superior image-level naturalness while providing enhanced feature-level generalization ability. Specifically, AdvCloak sequentially optimizes the generative adversarial networks by employing a two-stage training strategy. This strategy initially focuses on adapting the masks to the unique individual faces via image-specific training and then enhances their feature-level generalization ability to diverse facial variations of individuals via person-specific training. To fully utilize the limited training data, we combine AdvCloak with several general geometric modeling methods, to better describe the feature subspace of source identities. Extensive quantitative and qualitative evaluations on both common and celebrity datasets demonstrate that AdvCloak outperforms existing state-of-the-art methods in terms of efficiency and effectiveness.