Investigating the Performance of Gesture-Based Input for Mid-Air Text Entry in a Virtual Environment: A Comparison of Hand-Up versus Hand-Down Postures

Abstract

Although the interaction technology for virtual reality (VR) systems has evolved significantly over the past years, the text input efficiency in the virtual environment is still an ongoing problem. We deployed a word-gesture text entry technology based on gesture recognition in the virtual environment. This study aimed to investigate the performance of the word-gesture text entry technology with different input postures and VR experiences in the virtual environment. The study revealed that the VR experience (how long or how often using VR) had little effect on input performance. The hand-up posture has a better input performance when using word-gesture text entry technology in a virtual environment. In addition, the study found that the perceived exertion to complete the text input with word-gesture text entry technology was relatively high. Furthermore, the typing accuracy and perceived usability for using the hand-up posture were obviously higher than that for the hand-down posture. The hand-up posture also had less task workload than the hand-down posture. This paper supports that the word-gesture text entry technology with hand-up posture has greater application potential than hand-down posture.

A Comparative Study of the Typing Performance of Two Mid-Air Text Input Methods in Virtual Environments

“…When utilized in VR, mid-air input methods necessitate users to control their upper limbs more precisely, which could prove challenging and exhausting. Although indirect interaction can reduce the accuracy of mid-air interaction, it also increases the diversity of interactive positions [ 52 , 53 , 54 ]. Hence, the efficiency of these VR mid-air input methods remains largely unexplored.…”