A Study of Student Learning Status Classification Based on the Detection of Key Objects within the Visual Field
Qiubo Huang, Yixuan Hua · 2020
In order to improve students' concentration in solitary learning, this paper proposes a method to detect students' learning status. The students are first photographed, and then a Faster RCNN model is used to classify the learning status of the students in the photos. In order to improve the classification accuracy, the system detects the 2D face key points and matches them with the 3D face model to calculate the head angle based on the conversion between coordinates to determine the visual field of the eyes. Based on the visual field, it is possible to detect key objects that students may be focusing on, such as books, computers, cell phones. Based on these key objects, it can improve the accuracy of classifying students' learning status. The system can better identify learning status and help guardians to manage the students.