Survey on Sound and Video Analysis Methods for Monitoring Face-to-Face Module Delivery

The objective of this work is to identify unobtrusive methodologies that allow the monitoring and understanding of the educational environment, during face to face activities, through capturing and processing of sound and video signals. It is a survey on application and techniques that exploit these two signals (sound and video) retrieved in classrooms, offices and other spaces. We categorize such applications based upon the high level characteristics extracted from the analysis of the low level features of the sound and video signals. Through the overview of these technologies, we attempt to achieve a degree of understanding the human behavior in a smart classroom, on behalf of the students and the teacher. Additionally, we illustrate open-research points for further investigation.