A joint cascaded framework for simultaneous eye detection and eye state estimation

作者：

Highlights：

• An effective cascade regression method for simultaneous eye detection and eye state estimation is proposed.

• Based on a cascade regression framework, it iteratively estimates the location of the eye and the probability of the eye being occluded by eyelid.

• The regression models are learned from combination of generated synthetic photorealistic and real eye images.

• Experimental results on benchmark database show that it outperforms other state-of-the-art methods both on eye detection and eye state estimation. And it achieves real time applications.

摘要

•An effective cascade regression method for simultaneous eye detection and eye state estimation is proposed.•Based on a cascade regression framework, it iteratively estimates the location of the eye and the probability of the eye being occluded by eyelid.•The regression models are learned from combination of generated synthetic photorealistic and real eye images.•Experimental results on benchmark database show that it outperforms other state-of-the-art methods both on eye detection and eye state estimation. And it achieves real time applications.

论文关键词：Eye detection,Eye state estimation,Learning-by-synthesis,Cascade regression framework

论文评审过程：Received 15 September 2016, Revised 3 December 2016, Accepted 15 January 2017, Available online 1 February 2017, Version of Record 10 February 2017.

论文官网地址：https://doi.org/10.1016/j.patcog.2017.01.023