GaitGuard: Towards Private Gait in Mixed Reality

Augmented/Mixed Reality (AR/MR) technologies usher in a new era of immersive, collective experiences, distinctly differentiating them from traditional mobile systems. As these technologies evolve, prioritizing privacy and security is critical. This paper centers on gait privacy, a distinctive biometric vulnerable to revealing sensitive data. We introduce GaitGuard, a real-time system to safeguard gait privacy within MR environments. GaitGuard leverages a multi-threaded framework to efficiently process video frames, incorporating dedicated modules for stream capture, body detection and tracking, and privacy mitigation. This study includes a user analysis involving 20 participants to evaluate the risk of gait information exposure captured by video feeds in MR devices. Through thorough examination, we provide a comparative assessment of different mitigation techniques, analyzing their impact on privacy, video quality, and system efficiency. Our results indicate that GaitGuard significantly diminishes identification risks by up to $68\%$, while sustaining a robust streaming frame rate of $29$ FPS and preserving video clarity. GaitGuard offers a real-time approach to support privacy in MR applications, delivering a holistic solution to mitigate gait information exposure without affecting user experience.