19th International Conference

Keynote Speakers

John Boyd (DHS-OBIM, U.S.)

Identity: Asset or Liability?


The keynote talk will discuss the use of identity to access resources and benefits (i.e., identity as an asset) or its concealment to maintain anonymity (i.e., identity as a liability). The concept will be examined in light of pandemic effects such as those resulting from COVID-19 and identify appropriate modalities and emerging technologies for handling identity in ways informed by these events.


Over a 30-year career in the U.S. Navy, John Boyd served as a submariner on active duty, and in a variety of assignments in Navy and Joint headquarters as a reserve officer. In private industry, he led operations in chemical plants across the country, program management in chemical, biological, radiological, and nuclear defense information systems, hand-held biometric device prototype development, as well as software development for a major program in the federal government. In the government, he recently served as the Director, Defense Biometrics and Forensics on the Secretary of Defense Staff.

John Boyd is the Assistant Director, Office of Biometric Identity Management (OBIM) for the Department of Homeland Security. Leading the Futures Identity team, Mr. Boyd’s responsibilities include overseeing science and technology activities related to biometrics and identity services as well as special projects involving interagency and international partners, industry, and academia.

Reza Derakhshani (ZOLOZ, U.S.)

Mobile Biometrics: Translating Research into Practice


How do we establish trusted identities? We are who we are (the way of natural authentication, a.k.a. biometrics), not something that we memorize (passwords) or something that we have (physical keys and tokens). With their historically unprecedented global penetration, mobile devices are now our de facto all-in-one point of entry to digital life; but they are also prone to being misplaced or stolen. With more critical functions such as financial transactions tied to our devices, the critical question is: who’s holding them? In this talk we will discuss existing and upcoming biometric technologies, including one based on vasculature seen in the white of the eyes along with the surrounding micro features (Eyeprint). We will also mention important trends such as privacy-preserving biometrics and anti-spoofing for seamless personal authentication, and their role in a securely connected society free of passwords and physical tokens, delivering safe, private, and convenient trust.


Dr. Reza R. Derakhshani is a Professor of Computer Science and Electrical Engineering at University of Missouri, Kansas City. He is also the consulting Chief Scientist and technology inventor at ZOLOZ (formerly EyeVerify), a Kansas City biometric startup that was acquired by Alibaba’s Ant Financial in 2016. He earned his Ph.D. and Master’s degrees in Computer and Electrical Engineering from West Virginia University. Dr. Derakhshani's research interests are in biometrics, computational imaging, and machine learning. His interdisciplinary research has brought about close collaborations with various researchers and institutions. His research projects include mobile biometric identification centered around vasculature seen on the white of the eye (commercially known as Eyeprint ID), presentation attack detection for ocular and face biometrics, hyperfocal biometrics, and non-obtrusive affective-cognitive analysis using postural and ocular dynamics. His work has been sponsored by private industry and various state and federal agencies through 45 research contracts, and has resulted in over a hundred publications and issued patents. Dr. Derakhshani’s work has been featured in numerous radio, TV, and web/print media pieces around the world.

Shiguang Shan (Chinese Academy of Sciences, China)

Towards Domain-invariant Face Presentation Attack Detection


Conventional face Presentation Attack Detection (PAD) methods fail to generalize well to new application scenarios, due to the discrepancy of spoofing face distribution among domains. While domain adaptation can be a choice to address the domain shift problem, improving the domain generalization ability of face anti-spoofing techniques is more appealing and become a new trend in this field. In this talk, I will introduce some of our recent efforts along this research direction. The basic ideas behind our methods include multi-domain disentangled feature learning, real/fake asymmetric adversarial leaning, and dual meta-learning network by leveraging the depth information.


Shiguang Shan received Ph.D. degree in computer science from the Institute of Computing Technology (ICT), Chinese Academy of Sciences (CAS), Beijing, China, in 2004. He has been a full Professor of this institute since 2010 and now the deputy director of CAS Key Lab of Intelligent Information Processing and also a Professor with the University of Chinese Academy of Sciences. His research interests include computer vision, pattern recognition, and machine learning. He has published more than 300 papers, with totally more than 18,000 Google scholar citations. He served as Area Chairs (or Senior PC) for many international conferences including ICCV11, ICASSP14, ICPR12/14/19, ACCV12/16/18, FG13/18/20, BTAS18, AAAI20, and CVPR19/20. And he was/is Associate Editors of several journals including IEEE T-IP, Neurocomputing, CVIU, and PRL. He was a recipient of the China’s State Natural Science Award in 2015, and the China’s State S&T Progress Award in 2005 for his research work.