Skip to main content

Best Poster Awards at International Joint Conference on Biometrics 2017

Two Papers won Best Poster Awards!

The Intelligent Forensics, Biometrics and Security lab ( at West Virginia University under the leadership of Dr. Afzel Noore has been recognized for innovative biometrics research and won two best poster awards at the International Joint Conference on Biometrics (IJCB) held in Denver in October 2017. This prestigious conference on biometrics combines two major biometrics research annual conferences, the IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS) and the IAPR International Conference on Biometrics (ICB). The blending of these two conferences every three years creates an exciting venue for presenting cutting-edge biometrics research of highest quality to biometrics research community worldwide. Dr. Mayank Vatsa and Dr. Richa Singh, Associate Professors at IIIT-Delhi and Adjunct Associate Professors at West Virginia University collaborated on the two award winning papers. 

CS PhD graduate students from the Lane Department of Computer Science and Electrical Engineering, Naman Kohli and Daksha Yadav, received the best poster award for their paper titled “Synthetic Iris Presentation Attack using iDCGAN”. This paper proposes a novel algorithm to generate synthetic iris images using iris deep convolutional generative adversarial network (iDCGAN). The algorithm integrates iris quality attributes to generate realistic looking synthetic iris images. Even commercial iris recognition systems find it difficult to discriminate these synthetically generated iris images from real iris images.

CS PhD graduate students Maneet Singh and Shruti Nagpal from IIIT-Delhi spent nine months at West Virginia University as visiting research scholars. They developed a novel supervised deep class-encoder algorithm to predict the gender and ethnicity by analyzing iris images. They received the best poster award at the International Joint Conference on Biometrics for their innovative research titled “Gender and Ethnicity Classification of Iris Images using Deep Class Encoder”. The use of gender and ethnicity as a soft biometric improves the iris recognition performance, reduces the computational time, and results in faster processing.

“I am extremely pleased that the international community of leading biometrics researchers and professionals have recognized the quality and creativity of research performed by these graduate students”, said Dr. Noore. As iris biometric technology permeates many devices and applications, cybersecurity and privacy are of paramount importance. The research performed by these students explores integrating user specific soft biometrics attributes to develop new iris recognition algorithms. Synthetically generated iris images will be valuable in developing new algorithms to detect and mitigate iris spoofing attacks or cyber-attacks.