AuraMask: An Extensible Pipeline for Developing Aesthetic Anti-Facial Recognition Image Filters

TL;DR

AuraMask introduces a new pipeline for creating aesthetic anti-facial recognition filters that are both effective against recognition systems and more acceptable to users, demonstrated through 40 filters and a large user study.

Contribution

The paper presents AuraMask, a novel method for designing aesthetically pleasing AFR filters that outperform prior techniques in effectiveness and user acceptance.

Findings

AuraMask filters match or surpass prior methods in adversarial effectiveness.

A user study with 630 participants shows higher acceptance for AuraMask filters.

40 aesthetic filters were generated emulating popular Instagram effects.

Abstract

Anti-facial recognition (AFR) image filters alter images in ways that are subtle to people but blinding to computer vision. Yet, despite widespread interest in these technologies to subvert surveillance, users rarely use them in practice -- because the ``subtle'' alterations are visible enough to conflict with users' self-presentation goals. To address this challenge, we propose AuraMask: a novel approach to creating AFR filters that are both adversarially effective and aesthetically acceptable. Using AuraMask, we produce 40 ``aesthetic'' filters that emulate popular ``one-click'' Instagram image filters. We show that AuraMask filters meet or exceed the adversarial effectiveness of prior methods against open-source facial recognition models. Moreover, in a controlled online user study ($N=630$) we confirm these filters achieve significantly higher user acceptance than prior methods. Lastly, we provide our AFR pipeline to the community for accelerated research in adversarially effective and aesthetically acceptable protections.