Abstract:

The classroom atmosphere interplay between students and teachers, which is there by conceived of multiple expectations, consensus, and behaviors. Research has indicated that positive and uplifting learning environments facilitate learning efficacy and vice versa. To specify, when exposed to rage and frustration, learners tend to be fully occupied by negative messages, resulting in dim recollections and poor mindsets. Consequently, once messages fail to be thoroughly delivered to the neo-cortex layer to be carefully and critically processed, instincts take control. To further facilitate learning and teaching efficiency, especially in remote teaching, this study aims to incorporate classroom atmosphere modules into smart classroom establishments, which functions to measure classroom atmosphere, receive instant feedbacks, and do roll calls. Three techniques are explored in this study; they are atmosphere analyses, learning conditions, and roll call modules. This study develops and implements a Kinect-based 3D gesture recognition and body language recognition and face expression recognition technology for a teaching response system for teacher self-efficacy and student performance. The developed system detects and tracks human hands, body skeleton and face landmark from the RGBD images captured by a Kinect sensor and recognizes human behavior. Furthermore, we have explored and developed Real-time Learning Tracking (RTL) techniques and are expecting to adopt big data to provide more teaching and learning friendly environments. The developed system is implemented on a Windows 10 laptop PC using C++ and OpenCV 3.1.0 library, and tested in ordinary classroom environment. Its performance demonstrates the overall average accuracy of around 90% in recognizing status of body and face behavior commands under various ambient lighting conditions.

Keywords:

Classroom atmosphere, facial recognition, Real-time Learning Tracking.