Home / Current Issue / Paper 1717015
Revolutionizing Attendance Face Recognition Systems
Subject area: Science,Engineering and Technology · Area of research: Deep Learning & Face Recognition
DOI: 10.64388/IREV9I10-1717015
Abstract
Reliable documentation of attendance remains a persistent operational burden in educational institutions and corporate settings alike. Conventional solutions — spanning handwritten roll calls to token-based card readers — are routinely compromised by proxy attendance, labour-intensive record keeping, and limited scalability. Growing dissatisfaction with these limitations motivates the development of intelligent, low-infrastructure alternatives rooted in contemporary AI techniques. This paper proposes an automated, touchless attendance capture system founded on a sequential three-layer deep learning pipeline. The detection layer employs a Multi-Task Cascaded Convolutional Network (MTCNN) to scan live video and isolate individual faces along with associated anatomical landmarks. The representation layer then processes the geometrically normalised face crops through a FaceNet model built on an Inception-ResNet-v1 backbone, producing a compact 128-dimensional feature vector per face via triplet-loss-driven metric learning. Finally, the decision layer applies a Support Vector Machine with a Radial Basis Function kernel to associate each embedding with a registered identity that was enrolled offline. Evaluation spanned two distinct data sources: the publicly accessible Labeled Faces in the Wild (LFW) benchmark alongside an institution-specific corpus drawn from 50 volunteers photographed across varied lighting environments. The pipeline achieved 98.7% accuracy on the local dataset and 99.1% on LFW, with per-frame processing completing in 120 milliseconds. The False Acceptance Rate was confined to 0.8%. A Flask-powered administrative dashboard facilitates live monitoring and automated report export.
Keywords
Touchless Attendance, Facial Identification, MTCNN, Facenet Embeddings, SVM Classifier, Deep Metric Learning, Real-Time Recognition, Edge Deployment, Biometric Authentication.
References
[1] A. K. Jain, A. Ross, and S. Prabhakar, "An introduction to biometric recognition," IEEE Trans. Circuits Syst. Video Technol., vol. 14, no. 1, pp. 4–20, Jan. 2004.
[2] D. Kumar, C. S. Rai, and S. Kumar, "A novel approach for attendance management system using RFID and biometric," Int. J. Comput. Appl., vol. 58, no. 2, pp. 18–23, Nov. 2012.
[3] M. Turk and A. Pentland, "Eigenfaces for recognition," J. Cognitive Neurosci., vol. 3, no. 1, pp. 71–86, Jan. 1991.
[4] P. N. Belhumeur, J. P. Hespanha, and D. J. Kriegman, "Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection," IEEE Trans. Pattern Anal. Mach. Intell., vol. 19, no. 7, pp. 711–720, Jul. 1997.
[5] T. Ahonen, A. Hadid, and M. Pietikainen, "Face description with local binary patterns: Application to face recognition," IEEE Trans. Pattern Anal. Mach. Intell., vol. 28, no. 12, pp. 2037–2041, Dec. 2006.
[6] R. Ahuja, A. Jain, and M. Sharma, "Face recognition based attendance system using support vector machine," Int. J. Adv. Res. Comput. Sci., vol. 8, no. 5, pp. 412–416, May 2017.
[7] Y. Taigman, M. Yang, M. Ranzato, and L. Wolf, "DeepFace: Closing the gap to human-level performance in face verification," in Proc. IEEE CVPR, Columbus, OH, USA, 2014, pp. 1701–1708.
[8] F. Schroff, D. Kalenichenko, and J. Philbin, "FaceNet: A unified embedding for face recognition and clustering," in Proc. IEEE CVPR, Boston, MA, USA, 2015, pp. 815–823.
[9] J. Deng, J. Guo, N. Xue, and S. Zafeiriou, "ArcFace: Additive angular margin loss for deep face recognition," in Proc. IEEE CVPR, Long Beach, CA, USA, 2019, pp. 4690–4699.
[10] K. Zhang, Z. Zhang, Z. Li, and Y. Qiao, "Joint face detection and alignment using multi-task cascaded convolutional networks," IEEE Signal Process. Lett., vol. 23, no. 10, pp. 1499–1503, Oct. 2016.
[11] G. Guo and N. Zhang, "A survey on deep learning based face recognition," Comput. Vis. Image Underst., vol. 189, pp. 1–7, Dec. 2019.
[12] R. Pandey, A. Kumar, and S. Singh, "Real-time face recognition based attendance system using FaceNet and KNN classifier," Int. J. Eng. Res. Technol., vol. 9, no. 6, pp. 231–236, Jun. 2020.
[13] Y. Kortli, M. Jridi, A. Al Falou, and M. Atri, "Face recognition systems: A survey," Sensors, vol. 20, no. 2, pp. 1–34, Jan. 2020.
[14] A. Dosovitskiy et al., "An image is worth 16×16 words: Transformers for image recognition at scale," in Proc. ICLR, Vienna, Austria, 2021, pp. 1–22.
How to cite this paper
@article{1717015,
author = {Ashish Kumar, Akash Dwivedi, Suresh Kumar Tiwari, Dr. Sanjay Pachauri},
title = {Revolutionizing Attendance Face Recognition Systems},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {10},
pages = {3598-3603},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1717015.pdf},
abstract = {Reliable documentation of attendance remains a persistent operational burden in educational institutions and corporate settings alike. Conventional solutions — spanning handwritten roll calls to token-based card readers — are routinely compromised by proxy attendance, labour-intensive record keeping, and limited scalability. Growing dissatisfaction with these limitations motivates the development of intelligent, low-infrastructure alternatives rooted in contemporary AI techniques. This paper proposes an automated, touchless attendance capture system founded on a sequential three-layer deep learning pipeline. The detection layer employs a Multi-Task Cascaded Convolutional Network (MTCNN) to scan live video and isolate individual faces along with associated anatomical landmarks. The representation layer then processes the geometrically normalised face crops through a FaceNet model built on an Inception-ResNet-v1 backbone, producing a compact 128-dimensional feature vector per face via triplet-loss-driven metric learning. Finally, the decision layer applies a Support Vector Machine with a Radial Basis Function kernel to associate each embedding with a registered identity that was enrolled offline. Evaluation spanned two distinct data sources: the publicly accessible Labeled Faces in the Wild (LFW) benchmark alongside an institution-specific corpus drawn from 50 volunteers photographed across varied lighting environments. The pipeline achieved 98.7% accuracy on the local dataset and 99.1% on LFW, with per-frame processing completing in 120 milliseconds. The False Acceptance Rate was confined to 0.8%. A Flask-powered administrative dashboard facilitates live monitoring and automated report export.},
keywords = {Touchless Attendance, Facial Identification, MTCNN, Facenet Embeddings, SVM Classifier, Deep Metric Learning, Real-Time Recognition, Edge Deployment, Biometric Authentication.},
month = {April},
doi = {https://doi.org/10.64388/IREV9I10-1717015}
}