Bringing "Virtual" to "Reality": Enhancing Security and Usability on VR System and Applications

With the rapid advancements in computer science, electronics, optics, and related fields, virtual reality (VR) gradually penetrates into our daily lives, and is predicted to become a core technology in the near future. Despite its potentials, however, existing designs and solutions for VR applications remain at the infant stage, introducing limited usability and efficiency for real-world users. Besides, the increasing prevalence of VR presents new security and privacy threats due to the vast amount of information stored in or accessible through VR devices. To bridge this gap, we exploit and combine techniques from computer science and human biology, as well as other related domains, to enhance security, usability, and efficiency of these novel applications.

In the dissertation, we investigate the emerging security and privacy threats of existing VR systems, propose novel mitigation schemes, and develop new techniques to improve user experience in emerging VR applications. Our contributions are mainly threefold. First, we introduce novel user authentication schemes on VR via a secure and convenient visual channel. Specifically, we leverage the customized blink patterns and the biometric pupil variations to identify legitimate users, which is deployable for commercial VR devices. We further enhance this work by exploiting the phenomenon of auditory-pupillary response and introducing an effort-free biometric authentication scheme for VR devices, which outperforms all state-of-the-art solutions. Second, we propose to harness users' ocular behaviors to enable accurate quality of experience (QoE) assessment for 360-degree videos, by modeling these cues into a graph and applying graph learning techniques to extract hidden information in predicting the QoE score. Third, we build a novel video recommender system for VR users leveraging additional insights from users' physiological signals to learn their preferences and interests and make corresponding recommendations.