LowKey: Leveraging Adversarial Attacks to Protect Social Media Users from Facial Recognition

Abstract

Facial recognition systems are increasingly deployed by private corporations, government agencies, and contractors for consumer services and mass surveillance programs alike. These systems are typically built by scraping social media profiles for user images. Adversarial perturbations have been proposed for bypassing facial recognition systems. However, existing methods fail on full-scale systems and commercial APIs. We develop our own adversarial filter that accounts for the entire image processing pipeline and is demonstrably effective against industrial-grade pipelines that include face detection and large scale databases. Additionally, we release an easy-to-use webtool that significantly degrades the accuracy of Amazon Rekognition and the Microsoft Azure Face Recognition API, reducing the accuracy of each to below 1%.

arXiv commentPublished as a conference paper at ICLR 2021

Cited by in corpus (12)

• Privacy Intelligence: A Survey on Image Privacy in Online Social Networks
• On Success and Simplicity: A Second Look at Transferable Targeted Attacks
• Adversarial Examples Make Strong Poisons
• Adversarial Image Color Transformations in Explicit Color Filter Space
• Preventing Unauthorized Use of Proprietary Data: Poisoning for Secure Dataset Release
• Ulixes: Facial Recognition Privacy with Adversarial Machine Learning
• Better Safe Than Sorry: Preventing Delusive Adversaries with Adversarial Training
• Technical Challenges for Training Fair Neural Networks
• Rethinking Image-Scaling Attacks: The Interplay Between Vulnerabilities in Machine Learning Systems
• Identification of Attack-Specific Signatures in Adversarial Examples
• Addressing Privacy Threats from Machine Learning
• Fairness Properties of Face Recognition and Obfuscation Systems