Towards Robust Classroom Attendance: A Comprehensive Evaluation of Face Detection and Recognition Models

Himani Trivedi; Hiren Patel; Ridham Patel; Krutika Patel; Nancy Patel

cs.CVSep 19, 2026

Towards Robust Classroom Attendance: A Comprehensive Evaluation of Face Detection and Recognition Models

Authors: Himani Trivedi, Hiren Patel, Ridham Patel, Krutika Patel, Nancy Patel

Organizations: Computer Engineering Department LDRP-ITR, Kadi Sarva Vishwavidyalaya Gandhinagar, Gujarat, India · VSITR, Kadi Sarva Vishwavidyalaya Kadi, Gujarat, India · Information Technology Department LDRP-ITR, Kadi Sarva Vishwavidyalaya Gandhinagar, Gujarat, India

Abstract

Manual attendance methods, such as paper or register-based systems, take a lot of time, can lead to errors, and are easy to falsify. Face recognition is more reliable, but it frequently struggles in classrooms because lighting and other conditions can vary. Face recognition datasets are designed for regulated environments and do not capture the actual challenges found in classrooms. To address this, a new face detection and recognition dataset, the Visage Face dataset, comprising 16,234 face samples, is proposed for the task of face detection and recognition. The photos are taken from different angles and under varying lighting conditions, with students showing a range of expressions, and some faces partly covered to reflect real-life situations. A YOLO-based system is used to detect faces and tested seven advanced face recognition models with thirteen configurations: LVFace, QCFace, FaceLiVTv2, TopoFR, EdgeFace, TransFace, and GhostFaceNets. Of these, FaceLiVTv2-M performed best, with 99.75% Top-1/Top-5 accuracy and an inference time of 6.459 ms. These results show that the Visage Face Dataset is a realistic and challenging benchmark for face recognition in classroom attendance.

Figures & tables

Fig. 1: Sample images from the VisageFace-Detection Dataset (original vs. augmented)

Specification	Detection Details	Recognition Details
Total Images	3,635	3,500
Dataset Type	Categorical	Categorical
Dataset Size	224.6 MB	145.7 MB
Image Dimension	640 $\times$ 640 pixels	224 $\times$ 224 pixels
Total Annotations	96,409	-
Image Format	JPG	JPEG

TABLE I: VisageFace Dataset Summary

Fig. 2: Examples from the proposed VFD dataset under varying classroom conditions, including lighting, occlusion, and student density

Fig. 3: Real-time face detection using the selected YOLOv8n model, with bounding boxes and confidence scores in a classroom setting.

Fig. 4: Comparative Graph of Face Detection Models

Fig. 5: Inference Time, Accuracy, and Model Size Comparison of Evaluated FR Models

Model	Version	Params	Top-1	Top-5	Inference (ms)
FaceLiVTv2	L	8.52	99.75	100	6.081
EdgeFace	Base	18.23	99.75	99.75	4.029
LVFace	ViT-B	113.83	99.75	100	5.784
TopoFR	R100	65.16	99.5	100	11.411
QCFace	IR100	65.16	99.5	99.75	11.6
TransFace	ViT-B	124.47	99.75	100	6.081

TABLE III: Comparison of FR Models on the VisageFace-Recognition Dataset

Face Recognition

Face recognition matches or identifies people from facial images, videos, and biometric representations across changing conditions.

See all 108 papers

Benchmarking Face Recognition without Real Faces
Jul 16, 2026Paweł Borsukiewicz, Daniele Lunghi, Wendkûuni C. Ouédraogo +2Privacy-Preserving ML Face Recognition
University of Luxembourg · Luxembourg
WIDER-FAIR: An Annotated Version of the WIDER-FACE Dataset for Fairness Evaluation
Jun 30, 2026Maxime Moussi, Benoît Ronval, Siegfried Nijssen +1Algorithmic Fairness Algorithmic Bias
UCLouvain, Louvain-la-Neuve, Belgium · UCLouvain, ICTEAM, Louvain-la-Neuve, Belgium · UCLouvain, ICTEAM, Louvain-la-Neuve, Belgium and KU Leuven, DTAI, Leuven, Belgium +1
EventFace: Event-Based Face Recognition via Structure-Driven Spatiotemporal Modeling
Apr 8, 2026Qingguo Meng, Xingbo Dong, Zhe Jin +1Face Recognition Event-Based Vision
State Key Laboratory of Opto-Electronic Information Acquisition and Protection Technology, the Anhui Provincial Key Laboratory of Secure Artificial Intelligence, Anhui Provincial International Joint Research Center for Advanced Technology in Medical Imaging, and the School of Artificial Intelligence, Anhui University, Hefei, China · Computer Vision Laboratory, University of Sassari, 07100 Sassari, Italy

Jul 16, 2026J/K move · Enter open · S save

Benchmarking Face Recognition without Real Faces

Paweł Borsukiewicz, Daniele Lunghi, Wendkûuni C. Ouédraogo, Jacques Klein+1

University of Luxembourg · Luxembourg

Privacy-Preserving ML Face Recognition Synthetic Data Evaluation

Open paper AskPDF arXiv

Synthetic face datasets have become effective enough to train face recognition models with accuracy rivaling that of models trained on real photographs. This progress sidesteps the ethical and legal burdens of collecting real biometric data, yet evaluation has not kept pace. Even studies that train entirely on synthetic images still rely on real-face benchmarks to measure performance, leaving the privacy problem only half solved. We ask whether synthetic datasets can replace real benchmarks for face recognition evaluation. We test 12 synthetic datasets against 7 established real benchmarks using 24 pre-trained models that span both convolutional and transformer architectures. Our evaluation covers biometric verification metrics, similarity score distributions, cross-model ranking consistency, and the underlying distributional properties of each dataset. Benchmarking fidelity varies widely across the synthetic candidates, but the two strongest, MorphFace and Vec2Face, reproduce the relative behavior of real benchmarks and reach agreement levels that fall within the natural disagreement already observed among the real benchmarks themselves. These results establish that well-constructed synthetic datasets can support reliable comparative evaluation for face recognition, moving the field closer to a fully synthetic and privacy-preserving pipeline for both training and benchmarking.

Abstract

Face Recognition

Explore similar work

Benchmarking Face Recognition without Real Faces

WIDER-FAIR: An Annotated Version of the WIDER-FACE Dataset for Fairness Evaluation

EventFace: Event-Based Face Recognition via Structure-Driven Spatiotemporal Modeling