Can Privacy Be Satisfying? On Improving Viewer Satisfaction for Privacy-Enhanced Photos Using Aesthetic Tranforms
Pervasive photo sharing in online social media platforms cancause unintended privacy violations when elements of animage reveal sensitive information. Prior studies have iden-tified image obfuscation methods (e.g., blurring) to enhanceprivacy, but many of these methods adversely affect view-ers’ satisfaction with the photo, which may cause people toavoid using them. In this paper, we study the novel hypothe-sis that it may be possible to restore viewers’ satisfaction by‘boosting’ or enhancing the aesthetics of an obscured image,thereby compensating for the negative effects of a privacytransform. Using a between-subjects online experiment, westudied the effects of three artistic transformations on imagesthat had objects obscured using three popular obfuscationmethods validated by prior research. Our findings suggestthat using artistic transformations can mitigate some nega-tive effects of obfuscation methods, but more exploration isneeded to retain viewer satisfaction.
Hasan, Rakibul, Yifang Li, Eman Hassan, et al. "Can Privacy Be Satisfying? On Improving Viewer Satisfaction for Privacy-Enhanced Photos Using Aesthetic Tranforms." In CHI Conference on Human Factors in Computing Systems Proceedings, May 4-9, 2019, Glasgow, Scotland. New York: ACM.
Privacy, Image obfuscation, Image filtering