2017/06/25 by Terence Brouns, Brouns, Terence
Computer Science · Medicine · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Gaze Tracking and Assistive Technology #Glaucoma and retinal disorders #Ocular Surface and Contact Lens
paper · pdf · doi:10.48550/arxiv.1706.08189
openalex publication_date 2017/06/25 · openalex created_date 2022/10/02 · openalex updated_date 2026/07/28
Video-based eye tracking is a valuable technique in various research fields.\nNumerous open-source eye tracking algorithms have been developed in recent\nyears, primarily designed for general application with many different camera\ntypes. These algorithms do not, however, capitalize on the high frame rate of\neye tracking cameras often employed in psychophysical studies. We present a\npupil detection method that utilizes this high-speed property to obtain\nreliable predictions through recursive estimation about certain pupil\ncharacteristics in successive camera frames. These predictions are subsequently\nused to carry out novel image segmentation and classification routines to\nimprove pupil detection performance. Based on results from hand-labelled eye\nimages, our approach was found to have a greater detection rate, accuracy and\nspeed compared to other recently published open-source pupil detection\nalgorithms. The program's source code, together with a graphical user\ninterface, can be downloaded at https://github.com/tbrouns/eyestalker\n