Home // eTELEMED 2021, The Thirteenth International Conference on eHealth, Telemedicine, and Social Medicine // View article

Gaze Calibration of Eye Trackers for Head-Mounted Displays Using Eye-Frontalization Process

Authors:
Katsuyoshi Hotta
Oky Dicky Ardiansyah Prima
Takashi Imabuchi

Keywords: VR; EyeTracking; Gaze Calibration; Gaze correction.

Abstract:
In order to simplify the eye-gaze calibration required for gaze measurement in Virtual Reality Head-Mounted Displays (VR-HMDs), we proposed a new eye gaze calibration method that combines eye-frontalization and single-point calibration. Deep Convolutional Neural Network was used for the frontalization of the eyes. Our method enables to estimate the gaze coordinates on the screen as soon as the user puts on the VR-HMD and to improve the accuracy of gaze measurement due to slight changes in the positional relationship between the VR-HMD and the face by updating the parameters of the gaze calibration. The accuracy of the proposed method was about 5 degrees for both eyes.

Pages: 6 to 10

Copyright: Copyright (c) IARIA, 2021

Publication date: July 18, 2021

Published in: conference

ISSN: 2308-4359

ISBN: 978-1-61208-872-3

Location: Nice, France

Dates: from July 18, 2021 to July 22, 2021