De-anonymizing VR Avatars using Non-VR Motion Side-channels

Virtual Reality (VR) technology offers an immersive audio-visual experience to users through which they can interact with a digitally represented 3D space (i.e., a virtual world) using a headset device. By (visually) transporting users from their physical world to realistic virtual spaces, VR systems enable interactive and true-to-life versions of traditional applications such as gaming, remote conferencing and virtual tourism. However, VR applications also present significant user-privacy challenges. This paper studies a new type of privacy threat targeting VR users which attempts to connect their activities visible in the virtual world to their physical state sensed in the real world. Specifically, this paper analyzes the feasibility of carrying out a de-anonymization or identification attack on VR users by correlating visually observed movements of users' avatars in the virtual world with some auxiliary data (e.g., motion sensor data from mobile/wearable devices) representing their context/state in the physical world. To enable this attack, the paper proposes a novel framework which first employs a learning-based activity classification approach to translate the disparate visual movement data and motion sensor data into an activity-vector to ease comparison, followed by a filtering and identity ranking phase outputting an ordered list of potential identities corresponding to the target visual movement data. A comprehensive empirical evaluation of the proposed framework is conducted to study the feasibility of such a de-anonymization attack.