Research Article
An Accurate and Cost-Effective Approach Towards Real-Time Eye Movement Angle Estimation
@INPROCEEDINGS{10.1007/978-3-031-06368-8_10, author={Yunfeng Zhu and Linkai Tao and Zheng Zeng and Hangyu Zhu and Chen Chen and Wei Chen}, title={An Accurate and Cost-Effective Approach Towards Real-Time Eye Movement Angle Estimation}, proceedings={Wireless Mobile Communication and Healthcare. 10th EAI International Conference, MobiHealth 2021, Virtual Event, November 13--14, 2021, Proceedings}, proceedings_a={MOBIHEALTH}, year={2022}, month={6}, keywords={Electrooculography Eye movement angle estimation Non-linear polynomial model}, doi={10.1007/978-3-031-06368-8_10} }- Yunfeng Zhu
Linkai Tao
Zheng Zeng
Hangyu Zhu
Chen Chen
Wei Chen
Year: 2022
An Accurate and Cost-Effective Approach Towards Real-Time Eye Movement Angle Estimation
MOBIHEALTH
Springer
DOI: 10.1007/978-3-031-06368-8_10
Abstract
In this paper, an electrooculography (EOG)-based eye movement angle estimation approach, including signal acquisition, pre-processing, outlier removal and modeling, is proposed. The eye movement angle estimation model is a data-driven approach that using a non-linear polynomial method. It offers a simple, analytical, accurate, and cost-effective solution for real-time and large-space eye movement angle estimation. The feasibility of the proposed model was validated on a realistic scenario across 18 subjects. Experimental results show the horizontal estimation error in angle is less than 3.5°. Compared with most of the existing methods with high computational complexity, the proposed model can provide comparable results with less computational consumption cost in a large-space eye movement angle estimation. Meanwhile, the proposed model can be easily deployed in the embedded platform or mobile device with limited computing power and limited storage space for real-time eye movement angle estimation.
