Please use this identifier to cite or link to this item:
For citation please use:
Main Title: A one-point calibration design for hybrid eye typing interface
Author(s): Zeng, Zhe
Neuer, Elisabeth Sumithra
Roetting, Matthias
Siebert, Felix Wilhelm
Type: Article
Abstract: We present an eye typing interface with one-point calibration, which is a two-stage design. The characters are clustered in groups of four characters. Users select a cluster by gazing at it in the first stage and then select the desired character by following its movement in the second stage. A user study was conducted to explore the impact of auditory and visual feedback on typing performance and user experience of this novel interface. Results show that participants can quickly learn how to use the system, and an average typing speed of 4.7 WPM can be reached without lengthy training. The subjective data of participants revealed that users preferred visual feedback over auditory feedback while using the interface. The user study indicates that this eye typing interface can be used for walk-up-and-use interactions, as it is easily understood and robust to eye-tracking inaccuracies. Potential areas of application, as well as possibilities for further improvements, are discussed.
Subject(s): eye movement
eye tracking
gaze interaction
eye typing
Issue Date: 24-Jul-2022
Date Available: 1-Aug-2022
Language Code: en
DDC Class: 004 Datenverarbeitung; Informatik
Journal Title: International Journal of Human-Computer Interaction
Publisher: Taylor & Francis
Publisher DOI: 10.1080/10447318.2022.2101186
EISSN: 1532-7590
ISSN: 1044-7318
TU Affiliation(s): Fak. 5 Verkehrs- und Maschinensysteme » Inst. Psychologie und Arbeitswissenschaft » FG Arbeits-, Ingenieur- und Organisationspsychologie
Appears in Collections:Technische Universit├Ąt Berlin » Publications

Files in This Item:
Zeng_etal_One-Point_2022.pdf   Until 2023-07-25

Format: Adobe PDF | Size: 1.1 MB
Download   Request a copy

Item Export Bar

This item is licensed under a Creative Commons License Creative Commons