Recent advances in biometrics based on biomedical information focus on improving multiview biometric object detection for anti-spoofing frauds. The paper introduces an improved ArcFace (I-AF) model that combines Convolutional Neural Networks (CNN) for feature extraction with RetinaFace for face detection and recognition. This hybrid approach enhances accuracy and robustness in multi-face scenarios, achieving 97% accuracy in real-time student data tracking in a classroom. The system detects and recognizes individual faces, mapping results to attendance records. The I-AF model is trained on labeled student data, significantly outperforming existing methods. Facial Recognition Technology (FRT) has grown significantly, with applications in financial services, transportation, and education. It involves detecting face positions, preprocessing images, and recognizing faces by matching with a database. FRT can improve healthcare services and reduce identity-related fraud. However, it faces challenges such as bias, accuracy issues, and security vulnerabilities like spoofing. The proposed system addresses these limitations by achieving high accuracy in multi-face scenarios, crucial for human-face authentication. The research highlights the need for reliable and secure identity verification methods, especially in light of increasing identity theft and fraud cases. The study contributes by analyzing the impact of specific encoding schemes on facial recognition efficiency and performance. It also presents a combination of Deep Learning Algorithms (DLAs), RetinaFace for multi-face detection, and I-AF for individual recognition, demonstrating high accuracy in crowded environments. This framework provides a robust and efficient solution for tracking humans, addressing challenges in identity verification and fraud prevention.Recent advances in biometrics based on biomedical information focus on improving multiview biometric object detection for anti-spoofing frauds. The paper introduces an improved ArcFace (I-AF) model that combines Convolutional Neural Networks (CNN) for feature extraction with RetinaFace for face detection and recognition. This hybrid approach enhances accuracy and robustness in multi-face scenarios, achieving 97% accuracy in real-time student data tracking in a classroom. The system detects and recognizes individual faces, mapping results to attendance records. The I-AF model is trained on labeled student data, significantly outperforming existing methods. Facial Recognition Technology (FRT) has grown significantly, with applications in financial services, transportation, and education. It involves detecting face positions, preprocessing images, and recognizing faces by matching with a database. FRT can improve healthcare services and reduce identity-related fraud. However, it faces challenges such as bias, accuracy issues, and security vulnerabilities like spoofing. The proposed system addresses these limitations by achieving high accuracy in multi-face scenarios, crucial for human-face authentication. The research highlights the need for reliable and secure identity verification methods, especially in light of increasing identity theft and fraud cases. The study contributes by analyzing the impact of specific encoding schemes on facial recognition efficiency and performance. It also presents a combination of Deep Learning Algorithms (DLAs), RetinaFace for multi-face detection, and I-AF for individual recognition, demonstrating high accuracy in crowded environments. This framework provides a robust and efficient solution for tracking humans, addressing challenges in identity verification and fraud prevention.