• Abhiraj Mohan Kadam
• Nirbhay Singh

Face registration and recognition are fundamental components of biometric systems used for security, attendance, and workplace analytics. However, data processing pipelines often suffer from degraded performance under real-world conditions such as masks and glasses, extreme illumination, and non-frontal poses. In this paper, we propose a real-time, multi-camera face registration and recognition framework that integrates YOLOv8 for face detection with a CNN-based behavioral filter that accounts for mask status, pose variations, and partial occlusions. To improve robustness, our method performs multi-view registration, feature alignment using facial landmarks, and embedding-based matching with dynamic thresholds. The proposed system operates in real-time, achieving ≥30 FPS across multiple camera streams, while incorporating privacy-preserving features such as on-device inference and optional data redaction. Experimental validation shows that our system maintains high recognition accuracy in challenging scenarios, improving upon limitations identified in prior safety-compliance vision systems and mask-detection attendance pipelines. This work demonstrates the feasibility of deploying scalable, privacy-aware, and high-accuracy face registration systems in workplace environments. Recent works have shown the potential of combining facial recognition, mask detection, and temperature measurement into a unified employee management platform for ensuring workplace safety during and beyond the COVID-19 pandemic.

Face Registration, Face Recognition, Multi-Camera Systems, YOLOv8, Convolutional Neural Networks (CNN), Behavioral Filtering, Real-Time Monitoring, Employee Tracking, Computer Vision, Deep Learning, Activity Recognition Workplace Surveillance

"MULTI-CAMERA REAL-TIME FACE REGISTRATION AND RECOGNITION WITH YOLOv8 AND CNN-BASED BEHAVIORAL FILTERING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 9, page no.c702-c704, September-2025, Available :http://www.jetir.org/papers/JETIR2509277.pdf

"MULTI-CAMERA REAL-TIME FACE REGISTRATION AND RECOGNITION WITH YOLOv8 AND CNN-BASED BEHAVIORAL FILTERING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 9, page no. ppc702-c704, September-2025, Available at : http://www.jetir.org/papers/JETIR2509277.pdf