/Student-behavior-detection-system

The student behavior detection system developed based on mmdetection framework adopts swin transformer as the backbone network, adopts feature pyramid FPN as the neck part, and adopts Deformable DETR as the detection head. It detects seven common classroom behaviors: looking up, reading, writing, standing up, talking, and group discussion.

Primary LanguagePythonApache License 2.0Apache-2.0

This repository is not active