Effectiveness and Users' Experience of Obfuscation as a Privacy-Enhancing Technology for Sharing Photos

Abstract

Current collaborative photo privacy protection solutions can be categorized into two approaches: controlling the recipient, which restricts certain viewers' access to the photo, and controlling the content, which protects all or part of the photo from being viewed. Focusing on the latter approach, we introduce privacy-enhancing obfuscations for photos and conduct an online experiment with 271 participants to evaluate their effectiveness against human recognition and how they affect the viewing experience. Results indicate the two most common obfuscations, blurring and pixelating, are ineffective. On the other hand, inpainting, which removes an object or person entirely, and avatar, which replaces content with a graphical representation are effective. From a viewer experience perspective, blurring, pixelating, inpainting, and avatar are preferable. Based on these results, we suggest inpainting and avatar may be useful as privacy-enhancing technologies for photos, because they are both effective at increasing privacy for elements of a photo and provide a good viewer experience.

Social Media and the Scourge of Visual Privacy

“…To protect visual privacy, these methods can be implemented before posting the content or after identifying private objects. Most privacy technology uses one or more of these five protection techniques: intervention [23,26], blind vision [2,4,12], secure processing [12,15], redaction [27][28][29][30], and data hiding [12,15,31].…”

“…On SMNs, the users' information and visual content can intentionally or unintentionally be shared even though there may be a privacy risk contained [1]. From visual content, attackers can extract textual information, including credit card numbers, social security numbers, place of residence, phone numbers and other information [1,4]. This content can consequently create an opportunity for "cyberbullying" of other users [3].…”

Can Privacy-Aware Lifelogs Alter Our Memories?

“…We found that blurring improves quantitiy of recalled details, but does not improve the accuracy of recall. Although blurring for obfuscation is commonly used in research [7,13] and industry (e.g., Google Street View), we cannot make claims about the impact of other techniques on memory recall. We plan to compare in the impact of obfuscation in explicit photo capturing contexts such as taking selfies.…”

“…This is particularly true for environmental lifelogging [5]. Previous work counteracted this problem by 1) deleting photos: this can be done post hoc or by preventing the capture of photos if privacy-sensitive situations are detected [16], or by 2) obfuscating the individuals, for example by inpainting them and replacing them by avatars [13]. The second approach is particularly promising as it protects the individuals' privacy while maintaining good utility and user experience [13].…”

“…Although previous works (e.g., [7,13]) investigated the impact of obfuscation on privacy protection and user experience, it remains unclear how obfuscation impacts the recall of memories. Closing this gap is crucial as recalling memories is one of the main motivations behind lifelogging.…”