Pluto: Motion Detection for Navigation in a VR Headset

TL;DR

This paper introduces Pluto, a motion detection system for VR headsets that improves navigation accuracy by recognizing agent motion states from accelerometer data, reducing drift during tracking failures.

Contribution

Pluto is a novel system that detects motion and stillness in VR headsets using accelerometer data, enhancing navigation robustness in challenging environments.

Findings

87% accuracy in motion state recognition

40% reduction in navigation drift during failures

Effective in environments with poor illumination

Abstract

Untethered, inside-out tracking is considered a new goalpost for virtual reality, which became attainable with advent of machine learning in SLAM. Yet computer vision-based navigation is always at risk of a tracking failure due to poor illumination or saliency of the environment. An extension for a navigation system is proposed, which recognizes agents motion and stillness states with 87% accuracy from accelerometer data. 40% reduction in navigation drift is demonstrated in a repeated tracking failure scenario on a challenging dataset.