Department of Computer Science
Clarkson University, Potsdam, NY 13699, USA
Abstract: This talk will focus on our research on the development of two large-scale synthetic fingerprint datasets for contact-based and contactless biometric systems using Generative Adversarial Networks (GANs). Large fingerprint datasets are important for training and evaluation of biometric systems but are time-consuming and expensive to collect and require strict privacy measures. Researchers have been exploring the use of synthetic fingerprint data to mitigate these concerns. Motivated by the increasing demand for high-quality biometric datasets, we describe the use of StyleGAN2-ADA, StyleGAN3, and CycleGAN to generate realistic synthetic fingerprints that replicate bonafide fingerprints in terms of biometric features while preserving privacy. The methodology includes the generation of synthetic live fingerprints across various age groups and spoof fingerprints created with materials such as latex, Ecoflex, and gelatin. The resulting datasets comprise 12,500 contact-based images and 50,000 contactless images. Privacy and uniqueness evaluations using VeriFinger and BOZORTH3 matchers revealed minimal overlap with real datasets, affirming both security and reliability.
BIO: Syed Konain Abbas received his B.S. degree in Computer Science from the University of the Punjab, Pakistan, and his M.S. degree in Computer Science from the University of Engineering and Technology, Lahore, Pakistan. He is currently pursuing a Ph.D. in Computer Science at Clarkson University, Potsdam, NY. His research focuses on biometric systems, specifically the development of synthetic fingerprint datasets using Generative Adversarial Networks (GANs).
Thursday, January 30, 2025, 12:15-1:15 pm, CAMP 194
Join Link: https://clarkson.zoom.us/j/94088483647?pwd=2bmoDHPYozdIWgj1SrdVbqQaQUMUiz.1
*Co-Sponsored by IEEE student branch and HKN