Focus Estimation in Academic Environments Using Computer Vision

Abstract

In this paper we propose a system capable of monitoring students' focus through cameras and using Computer Vision algorithms. Experimental results show that our system is capable of identifying students and tracking their focus during a class. At the end of the class, the system outputs graphical feedback to teachers regarding the average level of students' focus. Moreover, it can identify lecture periods in which students were less watchful and the corresponding topics that potentially need extra focus. In this paper we start by presenting the architecture of the system, followed by results obtained both during a small-group workshop and a classroom with a large number of attending students. The main goal of this work is to contribute to the transformation of the classroom as a sensing environment, providing information to both teachers and students about their engagement during the class.