Harshvardhan

About the Project

Current SMS-OTP and Push notification based two-factor authentication solutions are prone to attacks and this is widely known in the research community. Thus, in this project we are developing a gesture based multi-factor authentication system. The user is expected to perform gestures with his smartphone in order to login into an account. I led this gesture recognition module and employed both machine learning and template matching based techniques parallelly to improve the robustness of the system. This ensures low false positive rates. We also aim at addressing real-world problems such as device variability, sensor drifts and processing delays in this work. We are in the process of filing a patent and a paper based on a part of this research is currently under review.

Read more about how the current two-factor Authentication technologies are faring on our blog - "How 2FAs are faring in the 2020s?".

Publication:

CAPTCHA 2FA

Jay Prakash, Harshvardhan C Takawale, Omkar Sathe and Tony Quek
Under blind review