{"title":"Eye Tracking and Measurement of Eye Rotation Using a Small Camera Installed roughly next to the Eye","authors":"K. Hoshino, Yuki Noguchi, Nayuta Ono","doi":"10.1109/HSI49210.2020.9142662","DOIUrl":null,"url":null,"abstract":"Because the eyeball is a sphere, the pupil center coordinate is distorted and this caused the issue of low eye tracking accuracy with the dark pupil eye tracking. The authors introduce a new eye tracking method that does not reduce the accuracy of estimates even when the pupil camera is installed roughly next to the human eye. In this method, the user is asked to look at a minimum of six points of light that include the corners of a pentagon that is slanted towards the inside corner of the eye and the center of gravity of that pentagon. The authors also introduce a method to measure eye rotation by using a camera fixed in the same manner as the aforementioned eye tracking. This method uses the characteristic images of the blood vessels in the whites of the eyeball. A template image is selected based on three criteria: First, the blood vessel must be thick and have good dark/light contrast; second, there must not be any external light sources reflected; and third, there must not be blood vessels with a similar shape nearby.","PeriodicalId":371828,"journal":{"name":"2020 13th International Conference on Human System Interaction (HSI)","volume":"116 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 13th International Conference on Human System Interaction (HSI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/HSI49210.2020.9142662","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Because the eyeball is a sphere, the pupil center coordinate is distorted and this caused the issue of low eye tracking accuracy with the dark pupil eye tracking. The authors introduce a new eye tracking method that does not reduce the accuracy of estimates even when the pupil camera is installed roughly next to the human eye. In this method, the user is asked to look at a minimum of six points of light that include the corners of a pentagon that is slanted towards the inside corner of the eye and the center of gravity of that pentagon. The authors also introduce a method to measure eye rotation by using a camera fixed in the same manner as the aforementioned eye tracking. This method uses the characteristic images of the blood vessels in the whites of the eyeball. A template image is selected based on three criteria: First, the blood vessel must be thick and have good dark/light contrast; second, there must not be any external light sources reflected; and third, there must not be blood vessels with a similar shape nearby.