AI-Driven Cloud-Based Real-Time Smart Attendance System with Edge Multi-Face Detection and AWS Rekognition

Author

Mrs. Ranjitha H G, Abhijith B K, Aman S M, Nithin G V, Supraj P


Abstract

 Verifying student lecture participation remains an indispensable statutory mandate across collegiate institutions for tracking pedagogical engagement and upholding academic rigor. Traditional roll-call registers and physical sign-in sheets squander nearly a quarter of instructional contact hours while remaining defenseless against proxy signatures, intentional record alteration, and clerical transcription errors. Although fixed biometric installations like optical fingerprint sensors and RFID card portals automate record capture, they cause severe entrance queuing congestion, create sanitary liabilities during infectious outbreaks, and require prohibitive capital outlays. This paper presents Smart Attendance, an autonomous edge-cloud hybrid platform that overcomes these systemic vulnerabilities through non-intrusive optical biometric logging.

The proposed architecture executes on-premise multi-face localization on an edge controller via an optimized YOLO convolutional neural network with boundary clamping and a 15% dynamic expansion margin, while delegating high-dimensional biometric vector embedding generation and approximate nearest-neighbor searching to Amazon Web Services (AWS) Rekognition and S3. To eliminate algorithmic misclassifications, a human-in-the-loop teacher reconciliation canvas enables faculty members to visually inspect candidate facial crops and resolve unmatched students before committing records to PostgreSQL. Verified attendance transitions immediately trigger asynchronous parent notifications via Twilio SMS and SMTP email workers. Evaluated across 120 students at GMIT, the platform attains 96.8% precision, 98.4% identification accuracy, and a 2.14-second end-to-end latency for 60-student cohorts, slashing administrative logging overhead by 86.0%.


Keywords

Smart Attendance, Multi-Face Detection, YOLO Architecture, AWS Rekognition, Edge-Cloud Synergy, Human-in-the-Loop Verification, Automated Notification, PostgreSQL.



Full Text:

Download Paper PDF


References


  1. Singh, T. Khan, A. Sharma, S. R, and S. Rajkumar, “Sustainable and Secure Real-Time Face Recognition and Notification System Using AWS Serverless Architecture,” 2025 IEEE International Conference on Computing, Power, and Communication Technologies (IC2PCT), Greater Noida, India, pp. 412–417, 2025, doi: 10.1109/IC2PCT64129.2025.10903412.
  2. Verma, S. Gupta, and P. Sharma, “Efficient Face Recognition-Based Attendance System,” 2024 IEEE International Conference on Emerging Trends in Electrical, Electronics and Computing Technologies (ICETEEC), Bengaluru, India, pp. 1–6, 2024, doi: 10.1109/ICETEEC61885.2024.10654321.
  3. K. Patel and A. Desai, “Attendance System Based on Deep Learning Face Recognition Without Queue,” International Journal of Intelligent Systems and Applications in Engineering, vol. 11, no. 2, pp. 345–353, 2023.
  4. N. Reddy and V. Kumar, “Smart Attendance Management System Based on Face Recognition Using CNN,” 2023 4th International Conference on Smart Electronics and Communication (ICOSEC), Trichy, India, pp. 884–889, 2023, doi: 10.1109/ICOSEC58147.2023.10275912.
  5. Ali, M. Hassan, and R. Khan, “IAAS: IoT-Based Automatic Attendance System With Photo Face Recognition in Smart Campus,” IEEE Internet of Things Journal, vol. 10, no. 18, pp. 16240–16252, 2023, doi: 10.1109/JIOT.2023.3271109.
  6. Mehta and K. Shah, “Real-Time Internet Based Attendance Using Face Recognition System,” 2023 IEEE Conference on Information and Communication Technology (CICT), Jabalpur, India, pp. 1–5, 2023, doi: 10.1109/CICT59885.2023.10452311.
  7. Sharma, A. Joshi, and N. Gupta, “Real-Time Face Recognition Attendance System Using Deep Learning and Cloud Integration,” 2023 IEEE 4th International Conference on Issues and Challenges in Intelligent Computing Techniques (ICIPTM), Ghaziabad, India, pp. 1–6, 2023, doi: 10.1109/ICIPTM57143.2023.10117892.
  8. Saraswat, M. Gupta, and A. Sharma, “Automatic Attendance System Using Facial Recognition Using the AdaBoost Algorithm,” in Proc. International Conference on Innovative Computing and Communication (ICICC 2024), Springer, Singapore, pp. 215–224, 2024, doi: 10.1007/978-981-97-0412-3_21.
  9. V. Dixit, A. Saxena, and S. Sharma, “Face Recognition-Based Attendance System,” in Proc. International Conference on Innovative Computing and Communication (ICICC 2024), Advances in Intelligent Systems and Computing, Springer, Singapore, pp. 189–198, 2024, doi: 10.1007/978-981-97-1542-6_19.
  10. Zhou and Y. Chen, “Real-Time Data Processing in Cloud Computing Environments,” IEEE, 2018.

 

 

Share your valuable work from Social Media Buttons