Accurate CNN-based pupil segmentation with an ellipse fit error regularization term

作者：

Highlights：

• Proposes Ellipse Fit Error as a regularization term for CNN-based pupil segmentation.

• Reports the best-known pupil detection results for ExCuSe & LPW datasets.

• Gives out GT Segmentation maps for the LPW dataset.

摘要

•Proposes Ellipse Fit Error as a regularization term for CNN-based pupil segmentation.•Reports the best-known pupil detection results for ExCuSe & LPW datasets.•Gives out GT Segmentation maps for the LPW dataset.

论文关键词：Pupil segmentation,Convolutional Neural Networks (CNN),UNet,Loss function,Regularization term

论文评审过程：Received 25 February 2021, Revised 23 July 2021, Accepted 28 September 2021, Available online 12 October 2021, Version of Record 20 October 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116004