DeepFakes for Privacy: Investigating the Effectiveness of State-of-the-Art Privacy-Enhancing Face Obfuscation Methods

Abstract

There are many contexts in which a person's face needs to be obfuscated for privacy, such as in social media posts. We present a user-centered analysis of the effectiveness of DeepFakes for obfuscation using synthetically generated faces, and compare it with state-of-the-art obfuscation methods: blurring, masking, pixelating, and replacement with avatars. For this, we conducted an online survey (N=110) and found that DeepFake obfuscation is a viable alternative to state-of-the-art obfuscation methods; it is as effective as masking and avatar obfuscation in concealing the identities of individuals in photos. At the same time, DeepFakes blend well with surroundings and are as aesthetically pleasing as blurring and pixelating. We discuss how DeepFake obfuscation can enhance privacy protection without negatively impacting the photo's aesthetics. CCS CONCEPTS• Human-centered computing → Empirical studies in HCI.

Shoulder Surfing through the Social Lens: A Longitudinal Investigation & Insights from an Exploratory Diary Study

“…This was followed by dimming filters, front camera previews, selective hiding, and low brightness (9.1% each). Participants also proposed modifications to the mechanisms, including blurring of faces in photos [27,31] and reduced notifications.…”

“…63.63% disagreed that having a mechanism will impact the relationship with the observer. Suggestions to improve selected mechanisms included fewer notifications and blurring of faces found in photos [27]. Overall, 36.36% of participants voiced to have the user interface as the controller of the mechanisms while 36.36% of participants wished to control the mechanism themselves.…”