Automated Classroom Engagement Evaluation using Machine Learning for 180 Degree Camera Environment.

Due to students' facial expressions recognition, the class instructor can analyze the students' understanding and concentration of the speech and lecture, which is a significant teaching impact evaluation. In order to find the key to the challenging of the considerable cost and low performance caused by hiring human analysts to follow classroom teaching impact, in this research, we proposed a new automated system that examines students' expressions. A deep learning model is employed in the system as a classifier to recognize each of the facial emotions. As a result, we obtain an average accuracy rate of 0.67% on FER2013. Our final result shows that the proposed algorithm can be used as an assistive system in real-time mode to analyze students' facial expressions and improve the efficiency of teaching evaluation.