The Leg-Swing: A Novel Seated Locomotion Interface for Improved Virtual Reality Immersion

Sung-Ha Lee, Ho-Taek Joo, Yunho Choi, Kyung-Joong Kim

Abstract: Locomotion in Virtual Reality (VR) is a crucial component that greatly affects user immersion. Traditional seated locomotion methods often use hand-based controllers, which reduce immersion due to inconsistencies between user and avatar movements. On the other hand, walking-based standing locomotion methods such as walking-in-place or redirected walking, while immersive, can cause physical fatigue and safety issues. This paper introduces a seated VR locomotion interface, the Leg-Swing, which aims to enhance immersion by reducing inconsistencies between user and avatar actions while walking compared to conventional seated locomotion methods, while simultaneously offering a less fatiguing experience compared to walking-based methods. The Leg-Swing allows users to navigate the VR environment by swinging their legs, which is inferred as locomotion action by pressure sensor data from the user’s hips using a deep learning model.