AuthNet: A Deep Learning Based Authentication Mechanism Using Temporal Facial Feature Movements (Student Abstract)

Authors

  • Mohit Raghavendra National Institute of Technology, Karnataka
  • Pravan Omprakash National Institute of Technology, Karnataka
  • Mukesh B R National Institute of Technology, Karnataka

Keywords:

Facial Recognition, Biometrics, Computer Vision, Applications Of AI, Information Extraction

Abstract

Deep learning algorithms are widely used to extend modern biometric authentication mechanisms in resource-constrained environments like smartphones, providing ease-of-use and user comfort, while maintaining a non-invasive nature. In this paper, an alternative is proposed, that uses both facial recognition and the unique movements of that particular face while uttering a password. The proposed model is language independent, the password doesn't necessarily need to be a set of meaningful words or numbers, and also, is a contact-less system. When evaluated on the standard MIRACL-VC1 dataset, the proposed model achieved a testing accuracy of 98.1%, underscoring its effectiveness.

Downloads

Published

2021-05-18

How to Cite

Raghavendra, M., Omprakash, P., & B R, M. (2021). AuthNet: A Deep Learning Based Authentication Mechanism Using Temporal Facial Feature Movements (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence, 35(18), 15873-15874. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/17933

Issue

Section

AAAI Student Abstract and Poster Program