ProViewer: A Proximity- and View-Aware Framework for Detecting UI Security Risks in Extended Reality Applications
TL;DR
This work proposes a hybrid detection framework that integrates event-driven object tracking with targeted, periodic analysis to efficiently identify UI integrity violations, providing an effective and scalable defense against critical XR UI security threats.
Abstract
The rapid adoption of Extended Reality (XR) platforms has introduced new user interface (UI) security risks, such as overlapping and invisible object vulnerabilities, stemming from the complex spatial integration of virtual content in users' perceptual environments. Existing detection techniques often require exhaustive, continuous scene scanning, resulting in substantial computational overhead and poor scalability for practical XR applications. In this work, we propose a hybrid detection framework that integrates event-driven object tracking with targeted, periodic analysis to efficiently identify UI integrity violations. Our approach restricts monitoring to objects that enter the user's proximity and field of view, applying focused spatial overlap and invisibility checks through collider and rendering state assessment. Evaluation on commodity XR hardware demonstrates that our framework achieves high detection accuracy with low runtime overhead, providing an effective and scalable defense against critical XR UI security threats.