2021/02/03 by Wolfgang Fuhl, Fuhl, Wolfgang · 1 citation
Computer Science · Medicine · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Gaze Tracking and Assistive Technology #Glaucoma and retinal disorders #Image and Video Processing (eess.IV) #Ocular Surface and Contact Lens #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2102.01921
openalex publication_date 2021/02/03 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
In this paper we present a new approach for pupil segmentation. It can be\ncomputed and trained very efficiently, making it ideal for online use for high\nspeed eye trackers as well as for energy saving pupil detection in mobile eye\ntracking. The approach is inspired by the BORE and CBF algorithms and\ngeneralizes the binary comparison by Haar features. Since these features are\nintrinsically very susceptible to noise and fluctuating light conditions, we\ncombine them with conditional pupil shape probabilities. In addition, we also\nrank each feature according to its importance in determining the pupil shape.\nAnother advantage of our method is the use of statistical learning, which is\nvery efficient and can even be used online.\nhttps://atreus.informatik.uni-tuebingen.de/seafile/d/8e2ab8c3fdd444e1a135/?p=%2FStatsPupil&mode=list\n