PrinType: Text Entry via Fingerprint Recognition

PubDate: January 2023

Teams: Tsinghua University

Writers: Zongjian Liu;Jieling He;Jianjiang Feng;Jie Zhou

PDF: PrinType: Text Entry via Fingerprint Recognition

Abstract

We present PrinType, a fingerprint recognition based typing technique for virtual reality. Different regions of fingers covered by friction ridges are assigned to different keys (i.e. letters, numbers, punctuation, or functions). Once the thumb-worn fingerprint sensor touches a finger, the contact region (and its key) is recognized by matching the current image with registered templates. Using only a small sensor, PrinType turns each segment of all fingers into a touchable key. We designed keyboard layouts corresponding to three interaction sub-spaces: whole finger keyboard, fingertip keyboard, and single-handed keyboard. A 12-person user study was conducted to evaluate the performance of different strategies. Our user evaluation showed that participants achieved an average of 29.56, 32.38, and 34.22 WPM with 0.79%, 0.20%, and 0.21% not corrected error rate in the three strategies. In addition, we provided a detailed analysis of various micro metrics to further understand user performance and technical characteristics. Overall, PrinType is favored by users for its usability, efficiency, and novelty.