The application of artificial intelligence to classroom attendance

Traditional class attendance requires the teacher to call the roll or other auxiliary devices, which requires manual intervention and occupies a lot of time for teachers and students. This paper introduces the improved YOLO8 algorithm to practice face detection accurately to complete class attendance. First, the global attention mechanism GcNet model is introduced into the algorithm to help the model focus on key information better. The performance of the model is optimized. Secondly, the loss function adopts WIoU method to enhance the detection ability of small faces. Finally, the improved target detection head realizes the detection of small faces. Experimental result show that the improved YOLO8 algorithm can improve mAP50 by about 6% compared with the unchanged YOLO8 algorithm. This mode can also be used for face recognition, which proves the practicability of this model.