Two-eye model-based gaze estimation from a Kinect sensor

Xiaolong Zhou; Haibin Cai; You Fu Li; Honghai Liu

doi:10.1109/ICRA.2017.7989194

Two-eye model-based gaze estimation from a Kinect sensor

Xiaolong Zhou, Haibin Cai, You Fu Li, Honghai Liu

City University of Hong Kong

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

419 Downloads (Pure)

Abstract

In this paper, we present an effective and accurate gaze estimation method based on two-eye model of a subject with the tolerance of free head movement from a Kinect sensor. To accurately and efficiently determine the point of gaze, i) we employ two-eye model to improve the estimation accuracy; ii) we propose an improved convolution-based means of gradients method to localize the iris center in 3D space; iii) we present a new personal calibration method that only needs one calibration point. The method approximates the visual axis as a line from the iris center to the gaze point to determine the eyeball centers and the Kappa angles. The final point of gaze can be calculated by using the calibrated personal eye parameters. We experimentally evaluate the proposed gaze estimation method on eleven subjects. Experimental results demonstrate that our gaze estimation method has an average estimation accuracy around 1.99°, which outperforms many leading methods in the state-of-the-art. © 2017 IEEE.

Original language	English
Title of host publication	2017 IEEE International Conference on Robotics and Automation
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1646-1653
Number of pages	8
ISBN (Electronic)	978-1509046331
ISBN (Print)	978-1509046348
DOIs	https://doi.org/10.1109/ICRA.2017.7989194
Publication status	Published - 24 Jul 2017
Event	2017 IEEE International Conference on Robotics and Automation - Singapore, Singapore Duration: 29 May 2017 → 3 Jun 2017

Conference

Conference	2017 IEEE International Conference on Robotics and Automation
Abbreviated title	ICRA
Country/Territory	Singapore
City	Singapore
Period	29/05/17 → 3/06/17

Access to Document

10.1109/ICRA.2017.7989194

ICRA2017
© 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Accepted author manuscript (Post-print), 1.03 MB

Cite this

@inproceedings{4c61c75859d54a09b29dce6a3c8b9d36,

title = "Two-eye model-based gaze estimation from a Kinect sensor",

abstract = "In this paper, we present an effective and accurate gaze estimation method based on two-eye model of a subject with the tolerance of free head movement from a Kinect sensor. To accurately and efficiently determine the point of gaze, i) we employ two-eye model to improve the estimation accuracy; ii) we propose an improved convolution-based means of gradients method to localize the iris center in 3D space; iii) we present a new personal calibration method that only needs one calibration point. The method approximates the visual axis as a line from the iris center to the gaze point to determine the eyeball centers and the Kappa angles. The final point of gaze can be calculated by using the calibrated personal eye parameters. We experimentally evaluate the proposed gaze estimation method on eleven subjects. Experimental results demonstrate that our gaze estimation method has an average estimation accuracy around 1.99°, which outperforms many leading methods in the state-of-the-art. {\textcopyright} 2017 IEEE.",

author = "Xiaolong Zhou and Haibin Cai and Li, {You Fu} and Honghai Liu",

year = "2017",

month = jul,

day = "24",

doi = "10.1109/ICRA.2017.7989194",

language = "English",

isbn = "978-1509046348 ",

pages = "1646--1653",

booktitle = "2017 IEEE International Conference on Robotics and Automation",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "2017 IEEE International Conference on Robotics and Automation, ICRA ; Conference date: 29-05-2017 Through 03-06-2017",

}

Zhou, X, Cai, H, Li, YF & Liu, H 2017, Two-eye model-based gaze estimation from a Kinect sensor. in 2017 IEEE International Conference on Robotics and Automation ., 7989194, Institute of Electrical and Electronics Engineers Inc., pp. 1646-1653, 2017 IEEE International Conference on Robotics and Automation, Singapore, Singapore, 29/05/17. https://doi.org/10.1109/ICRA.2017.7989194

TY - GEN

T1 - Two-eye model-based gaze estimation from a Kinect sensor

AU - Zhou, Xiaolong

AU - Cai, Haibin

AU - Li, You Fu

AU - Liu, Honghai

PY - 2017/7/24

Y1 - 2017/7/24

N2 - In this paper, we present an effective and accurate gaze estimation method based on two-eye model of a subject with the tolerance of free head movement from a Kinect sensor. To accurately and efficiently determine the point of gaze, i) we employ two-eye model to improve the estimation accuracy; ii) we propose an improved convolution-based means of gradients method to localize the iris center in 3D space; iii) we present a new personal calibration method that only needs one calibration point. The method approximates the visual axis as a line from the iris center to the gaze point to determine the eyeball centers and the Kappa angles. The final point of gaze can be calculated by using the calibrated personal eye parameters. We experimentally evaluate the proposed gaze estimation method on eleven subjects. Experimental results demonstrate that our gaze estimation method has an average estimation accuracy around 1.99°, which outperforms many leading methods in the state-of-the-art. © 2017 IEEE.

AB - In this paper, we present an effective and accurate gaze estimation method based on two-eye model of a subject with the tolerance of free head movement from a Kinect sensor. To accurately and efficiently determine the point of gaze, i) we employ two-eye model to improve the estimation accuracy; ii) we propose an improved convolution-based means of gradients method to localize the iris center in 3D space; iii) we present a new personal calibration method that only needs one calibration point. The method approximates the visual axis as a line from the iris center to the gaze point to determine the eyeball centers and the Kappa angles. The final point of gaze can be calculated by using the calibrated personal eye parameters. We experimentally evaluate the proposed gaze estimation method on eleven subjects. Experimental results demonstrate that our gaze estimation method has an average estimation accuracy around 1.99°, which outperforms many leading methods in the state-of-the-art. © 2017 IEEE.

U2 - 10.1109/ICRA.2017.7989194

DO - 10.1109/ICRA.2017.7989194

M3 - Conference contribution

SN - 978-1509046348

SP - 1646

EP - 1653

BT - 2017 IEEE International Conference on Robotics and Automation

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2017 IEEE International Conference on Robotics and Automation

Y2 - 29 May 2017 through 3 June 2017

ER -

Two-eye model-based gaze estimation from a Kinect sensor

Abstract

Conference

Access to Document

Fingerprint

Cite this