International Journal of Scientific & Technology Research

Home About Us Scope Editorial Board Blog/Latest News Contact Us
10th percentile
Powered by  Scopus
Scopus coverage:
Nov 2018 to May 2020


IJSTR >> Volume 9 - Issue 10, October 2020 Edition

International Journal of Scientific & Technology Research  
International Journal of Scientific & Technology Research

Website: http://www.ijstr.org

ISSN 2277-8616

A Comparative Analysis Of Face Recognition Models On Masked Faces

[Full Text]



Yalavarthi Bharat Chandra, Gouru Karthikeya Reddy





Face recognition systems are among the widely used biometrics in fields such as surveillance, access controls, attendance, forensics and other security purposes. Due to current Covid-19 crisis almost everyone can be seen wearing a mask in public. This change can be very challenging for existing facial recognition systems and can make them less effective. A face mask covers significant portion of the face making facial recognition systems having less face features to recognize and on top of that face masks can also add significant noise to the image. As wearing face masks is going to be new normal it is important to understand and study how current state of art face recognition models perform in recognizing masked faces. We did a comparative analysis on four state of art deep learning models which are widely used in this field 1)VGGFace 2)FaceNet 3)OpenFace 4)DeepFace. The analysis is made on face verification task on RMFRD dataset which is largest real world masked face dataset available. We compared the models on various metrics like error rate, accuracy ,precision, verification time.



[1] Mathworks. [Online]. Available: https://in.mathworks.com/ [Accessed: January 2, 2019].
[2] T. Sabharwal and R. Gupta, “Facial Marks for Enhancing Facial Recognition after Plastic Surgery,” in Proc. of the 5th International Conference on Computing for Sustainable Global Development (INDIACom-2018)”, 14th - 16th March, 2018, Delhi, India.
[3] Poses: U. Scherhag, R. Raghavendra, K.B. Raja, M. Gomez-Barrero, C. Rathgeb, C. Busch, On the vulnerability of face recognition systems towards morphed face attacks, in 2017 5th International Workshop on Biometrics and Forensics (IWBF) (April 2017), pp. 1–6VGGface
[4] Omkar, Parkhi, Andrea Vedaldi, and Andrew Zisserman. "Deep face recognition." BMVC. Vol. 1. No. 3. 2015.
[5] Deep face recognition using imperfect facial dataAuthor links open overlay panelAliElmahmudiHassanUgailMasked Face Recognition Dataset and ApplicationFacenet
[6] Face Recognition using FaceNet (Survey, Performance Test, and Comparison)
[7] FaceNet: A Unified Embedding for Face Recognition and Clustering
[8] OpenFacehttps://algorithmia.com/blog/understanding-facial-recognition-openface
[9] DeepFace https://pypi.org/project/deepface/