PrivacyScout: Assessing Vulnerability to Shoulder Surfing on Mobile Devices

One approach to mitigate shoulder surfing attacks on mobile devices is detect the presence of a bystander using phone’s front-facing camera. However, person’s face in camera’s field view does not always indicate an attack. To overcome this limitation, novel data collection study (N=16), we analysed influence three viewing angles and four distances success attacks. In contrast prior works that mainly focused user authentication, investigated common types content susceptible surfing: text, photos, PIN authentications. We show vulnerability text photos depends observer’s location relative device, while authentications are vulnerable independent observation location. then present PrivacyScout – method predicts shoulder-surfing risk based visual features extracted from as captured by Finally, evaluations our demonstrate method’s feasibility assess attack more accurately.