Student Eye Gaze Tracking During MOOC Teaching
Fangfang Yang, Zengru Jiang, Chaojun Wang, Yaping Dai, Zhiyang Jia, Kaoru Hirota · 2018
In order to solve the problem that the teacher cannot monitor students e-learning status and evaluate the students e-learning effect during MOOC teaching, in this paper, we capture the student learning images by computer camera and adopt the "Three Angles Synthesis" to judge whether the student is watching the teaching video online. First locate the center of the iris by the Average Feature Point method to calculate iris deflection angle and the distance from the iris center to the eye center; and detect the face key feature points(eye, nose, mouth, etc.) by calling face key point detector in Dlib to determine the face rotation angel; then extract the student target body by Maximum Connected Domain Moment Characteristic method to calculate the student body tilt angle; last determine whether the three angles meet the giving conditions, further judge the student is watching the teaching video online or not.