Towards Safer Mixed Reality with Auditory Warnings for Uncontrolled Indoor Settings
Nasim Ahmed, Taeyeong Choi, Xinyue Zhang, Sahidul Islam, Rifatul Islam · 2025
Mixed Reality (MR) applications are increasingly gaining popularity across various sectors, making user safety crucial, as distractions from augmented and virtual content can cause injuries. Native safety features such as user-defined Guardian boundaries offer limited protection in controlled indoor settings. While limited research has been done to address safety in dynamic uncontrolled environments, we propose a novel method providing verbal feedback to help users safely navigate uncontrolled indoor settings. Our proposed approach achieved a mean accuracy of 93.64% with an average latency of 380 ms; thus demonstrating promising results for ensuring safety in MR interactions.