SCB-dataset: A Dataset for Detecting Student Classroom Behavior

Using deep learning methods to detect the classroom behaviors of both students and teachers is an effective way to automatically analyze classroom performance and enhance teaching effectiveness. Then, there is still a scarcity of publicly available high-quality datasets on student-teacher behaviors. Based on the SCB-Dataset3 we proposed previously, we have introduced a larger, more comprehensive, and higher-quality dataset of student-teacher classroom behaviors, known as SCB-Dataset5. Our dataset comprises 7428 images and 106830 labels across 20 classes: hand-raising, read, write, bow head, turn head, talk, guide, board writing, stand, answer, stage interaction, discuss, clap, yawn, screen, blackboard, teacher, leaning on the desk, using the phone, using the computer. We evaluated the dataset using the YOLOv7 series of algorithms We believe that SCB-Dataset5 can provide a solid foundation for future applications of artificial intelligence in education. Our SCB-Dataset5 can be downloaded at the following lhttps://github.com/Whiffe/SCB-dataset