Fairness Properties of Face Recognition and Obfuscation Systems
This addresses privacy and fairness concerns for individuals affected by automated face recognition, highlighting an incremental issue in existing obfuscation methods.
The paper investigates demographic fairness in face obfuscation systems, which use evasion attacks on metric embedding networks to protect privacy, and finds that these systems have reduced utility for minority groups due to clustering of face embeddings by demographic.
The proliferation of automated face recognition in the commercial and government sectors has caused significant privacy concerns for individuals. One approach to address these privacy concerns is to employ evasion attacks against the metric embedding networks powering face recognition systems: Face obfuscation systems generate imperceptibly perturbed images that cause face recognition systems to misidentify the user. Perturbed faces are generated on metric embedding networks, which are known to be unfair in the context of face recognition. A question of demographic fairness naturally follows: are there demographic disparities in face obfuscation system performance? We answer this question with an analytical and empirical exploration of recent face obfuscation systems. Metric embedding networks are found to be demographically aware: face embeddings are clustered by demographic. We show how this clustering behavior leads to reduced face obfuscation utility for faces in minority groups. An intuitive analytical model yields insight into these phenomena.