Generating Master Faces for Dictionary Attacks with a Network-Assisted Latent Space Evolution

TL;DR

This paper introduces a method to generate master face images using a neural network-guided evolutionary algorithm in StyleGAN's latent space, achieving high impersonation success across multiple face recognition systems.

Contribution

It presents a novel neural network-assisted evolutionary approach for creating master faces, significantly improving attack efficiency without additional identity data.

Findings

Over 40% identity coverage on LFW with fewer than 10 master faces

Neural network guidance enhances evolutionary search efficiency

Effective across multiple deep face recognition systems

Abstract

A master face is a face image that passes face-based identity-authentication for a large portion of the population. These faces can be used to impersonate, with a high probability of success, any user, without having access to any user-information. We optimize these faces, by using an evolutionary algorithm in the latent embedding space of the StyleGAN face generator. Multiple evolutionary strategies are compared, and we propose a novel approach that employs a neural network in order to direct the search in the direction of promising samples, without adding fitness evaluations. The results we present demonstrate that it is possible to obtain a high coverage of the LFW identities (over 40%) with less than 10 master faces, for three leading deep face recognition systems.