Light-Occlusion Text Entry in Mixed Reality

Abstract

Text entry is a recurring task in mixed reality (MR) applications, and the ability of eyes-free text entry methods to allow users to enter text without focusing on the input device is ideal and compelling. However, existing eyes-free text entry methods leave much to be desired regarding efficiency and accuracy. In this paper, we propose a new light-occlusion text entry method in MR environment that uses dual thumb typing on a touchscreen. We design a partially visible keyboard as visual feedback to improve user performance. In addition, we optimize the underlying keyboard by collecting eyes-free typing data through a user study. The results show that our method has high typing speed, low error rate, and is very novice-friendly. After a short training period, the average typing speed of the novice group can reach 26.23 WPM (words per minute), while the average typing speed of the potential expert group can reach 30.62 WPM.