Attention analysis in e-learning environment using a simple web camera

Abstract

A real-time non-intrusive attention tracking system using a simple web camera is proposed in this paper. This system is scale and rotation invariant and tolerant to blink related false attention state classifications. Attention states of the students are classified into three: attentive, sleepy and disappeared. A simple geometric model for eye corners detection is proposed. Active and passive attention tracking experiments are conducted with a 54 minutes video lecture as the e-learning session content. Experimental results show that the proposed system clearly discriminates the attention states of the student participated in the E-learning session. The execution time of the proposed algorithm is 10 milliseconds per image frame. The proposed system is highly suitable for real-time applications.