Evaluation of temporal stability of eye tracking algorithms using webcams

作者：

Highlights：

• For analysis of the movements of the eyes, the results should be precise and stable.

• A layered framework for pupil detection is proposed and implemented.

• A measure for temporal stability of pupil detection algorithms is proposed.

• The GPF algorithm is the most stable with less noisy sources.

• The Gradient and Combined algorithms are temporally stable with noisy sources.

摘要

•For analysis of the movements of the eyes, the results should be precise and stable.•A layered framework for pupil detection is proposed and implemented.•A measure for temporal stability of pupil detection algorithms is proposed.•The GPF algorithm is the most stable with less noisy sources.•The Gradient and Combined algorithms are temporally stable with noisy sources.

论文关键词：Eye tracking,Webcam,Temporal stability,Noise,Relative error,Efficiency

论文评审过程：Received 11 March 2016, Revised 20 June 2016, Accepted 19 July 2016, Available online 22 July 2016, Version of Record 27 July 2016.

论文官网地址：https://doi.org/10.1016/j.eswa.2016.07.029