Distribution Discrepancy Maximization for Image Privacy Preserving

TL;DR

This paper introduces a novel image obfuscation method that maximizes distribution discrepancy between original and encrypted images, effectively defending against deep recognition attacks by leveraging a collaborative training scheme.

Contribution

It proposes a new distribution discrepancy maximization approach for image privacy, with a collaborative training scheme and theoretical proof of its effectiveness.

Findings

Significant accuracy reduction on FaceScrub, Casia-WebFace, and LFW datasets.

Effective defense against deep recognition model attacks.

Outperforms traditional obfuscation methods.

Abstract

With the rapid increase in online photo sharing activities, image obfuscation algorithms become particularly important for protecting the sensitive information in the shared photos. However, existing image obfuscation methods based on hand-crafted principles are challenged by the dramatic development of deep learning techniques. To address this problem, we propose to maximize the distribution discrepancy between the original image domain and the encrypted image domain. Accordingly, we introduce a collaborative training scheme: a discriminator $D$ is trained to discriminate the reconstructed image from the encrypted image, and an encryption model $G_e$ is required to generate these two kinds of images to maximize the recognition rate of $D$, leading to the same training objective for both $D$ and $G_e$. We theoretically prove that such a training scheme maximizes two distributions' discrepancy. Compared with commonly-used image obfuscation methods, our model can produce satisfactory defense against the attack of deep recognition models indicated by significant accuracy decreases on FaceScrub, Casia-WebFace and LFW datasets.