Automated Multi Face Recognition and Identification using Facenet and VGG-16 on Real-World Dataset for Attendance Monitoring System

The goal of this study is to create a facial recognition-based automatic attendance tracking system. We suggested a Convolutional Neural Network (CNN) method for this system's real-time recognition and identification of many faces. For face embedding and identification, we use the FaceNet model, which has demonstrated outstanding performance in recent research. The suggested method uses a camera to take pictures of people, then extracts face traits and compares them to the database of registered users. The recognition of registered individuals is used to record attendance. The proposed system's accuracy is 75%. For single faces, the accuracy of Facenet and VGG-16 are 99.20% and 51.30% respectively.The system can be used in a number of places, including companies, colleges, and schools, where keeping track of attendance is crucial for administrative reasons. The automated attendance monitoring process of the suggested system can greatly reduce human error and save time. Furthermore, recognizing unwanted visitors helps improve the facility's security and safety.