As cameras become increasingly pervasive, video can reveal far more than what is directly visible. Seemingly innocuous visual cues, such as subtle hand movements or blurred text silhouettes, can expose sensitive information about a user’s actions and activities. In this talk, I will present our recent research on emerging video-based privacy attacks and defenses. First, I will show how video recordings of hand and controller movements can be exploited to infer PINs entered on virtual keyboards in Virtual Reality, even though the keyboard itself is invisible to the attacker. I will then demonstrate how blurry video recordings can reveal what a person is watching by exploiting the spatiotemporal patterns of subtitle silhouettes, even when the video content and subtitle text are unreadable. I will also discuss practical defenses against these attacks and conclude with broader implications and future directions for protecting user privacy in increasingly camera-rich environments.

Song Fang is an Associate Professor in the Knight Foundation School of Computing and Information Sciences at Florida International University. Previously, he was an Assistant Professor (2018-2024) and Associate Professor (2024-2026) at the University of Oklahoma, where he held the University's William H. Barkow Presidential Professorship. His research interests include cyber-physical systems and IoT security, wireless and mobile system security, and the intersection of machine learning and security. His research has appeared in premier journals and conferences, including IEEE Transactions on Dependable and Secure Computing, IEEE Transactions on Information Forensics and Security, ACM Transactions on Privacy and Security, IEEE Transactions on Mobile Computing, CCS, NDSS, MobiCom, MobiSys, INFOCOM, PETS, RAID, and ACSAC.