Toward Predicting User Waist Location From VR Headset and Controllers Through Machine Learning

PubDate: November 2021

Teams: University of Southern California

Writers: Adityan Jothi；Powen Yao；Andrew Zhao；Mark Miller；Sloan Swieso；Michael Zyda

PDF: Toward Predicting User Waist Location From VR Headset and Controllers Through Machine Learning

Abstract

Commercial VR Headsets typically include a headset and two motion controllers. From this VR setup, we have access to the user’s head and hands, but lack information about other parts of the user’s body without using additional equipment. Accurate position of other body parts such as the waist would expand the user’s interaction space. In this paper, we describe our efforts at using machine learning to predict the position and rotation of the user’s waist using only the headset and two motion controllers with an additional tracker at the waist for training.