– BIAS/EQUITY
– EXPLAINABLE AI
– VULNERABILITIES
– GENERATIVE AI
– TRANSPARENCY
– EXPLAINABILITY
– SECURITY
– PERFORMANCE
With our interdisciplinary group of faculty, researchers and students, CITeR is the only National Science Foundation (NSF) Industry/University Cooperative Research Center (I/UCRC) focusing on serving its affiliates in the rapidly growing areas of identity and biometrics.
CITeR Affiliates
ACV Auctions
Athena Sciences
Aware, Inc
CVS Health
Defense Forensics and Biometrics Agency (DFBA)
Defense Research and Development Canada (DRDC)
DHS Office of Biometric Identity Management (OBIM)
23
21
master’s and PhD students currently engaged in CITeR research
17
faculty engaged in research and teaching biometrics
webinars featuring results from recently completed CITeR projects
DHS Science & Technology (DHS S&T)
FBI Criminal Justice Information Services (CJIS)
FBI Operational Technology Division (OTD)
General Services Administration (GSA)
Home Team Science and Technology Agency (HTX), Singapore (2024)
IDEMIA
Ingenium
iProov
Metalenz (2024)
National Security Agency (NSA) (2024)
Neo Auth (2024)
Oak Ridge National Laboratory
Precise Biometrics
Public Safety Canada
Qualcomm
SICPA
Synolo™ Biometrics
Tech5
Thales
Tools for Humanity (2024)
U.S. Army Criminal Investigation Laboratory (USACIL)
Veridium
Message From the Director
The use of biometric recognition and artificial intelligence (AI) to determine legitimacy of identity for commercial and government systems is ever expanding.
The proliferation of PII has brought widespread concerns around bias, surveillance, political persecution, and information exploitation. The Center of Identification Technology Research (CITeR) serves to address the needs of society during rapid technological change while ensuring the guardrails around technology are grounded in an understanding of the science.
As our organization transitions into the third phase of the NSF IUCRC journey, after 20 years in existence, I’m reflecting on the relationships with Industry and Government that fuel CITeR — see the current affiliate list to the left. Membership in CITeR allows organizations to influence and guide the CITeR research portfolio. This cooperative model between Affiliates and the University researchers allows for the advancement of the fundamental body of identity research while simultaneously supporting the research platform of our affiliate members.
The CITeR Affiliate community is composed of key technical representatives from the member organizations. Since its establishment, CITeR has been fortunate to have informed, engaged affiliate members. This couldn’t be more true today — significant research cycle engagement, hosting CITeR workshops, involvement in CITeR outreach STEM events, and active engagement with CITeR researchers and student populations. This active engagement strengthens the CITeR community and promotes mutually beneficial outcomes for all.
In Phase 3, CITeR will continue efforts to provide value to our current members and look to establish new verticals, including Banking, Healthcare, and Social Good — expanding the research portfolio to benefit all. UN Goals and Presidential orders surrounding the efforts to create electronic identities for all and the recognition that societies must support the safe, secure and fair use of AI through privacy-enhancing technologies support the value of the CITeR research portfolio. We look forward to helping our affiliate community solve their research challenges in their portfolios.
Stephanie Schuckers, Director of
the Center for Identification Technology Research (CITeR)
A collection of CITeR face masks and custom figurines use in CITeR Presentation Attack Detection Research.
MD Jahangir Alam Khondkar shares their CITeR funded research as part of a student research event held at the CITeR Fall 2023 Program Review.
Dennis Fedorishin (UB) receives the CITeR Spring 2024 Best Poster Award from directors Matt Valenti (WVU) and Srirangaraj Setlur (UB).
Research Highlights
- WEST VIRGINIA UNIVERSITY
- UNIVERSITY AT BUFFALO
- CLARKSON UNIVERSITY
Advancing Noncontact Fingerprint Recognition
Noncontact fingerprint recognition is poised for growth given the cost, hardware requirements and logistical challenges of traditional fingerprint systems. The pervasiveness of cell phone ownership propels exploration of noncontact fingerprint in certain use cases. CITeR Researchers are researching in this area, addressing contactless fingerprint interoperability with legacy contact fingerprints and evaluating potential sources of differential performance.
RESEARCH DIRECTIONS
• Develop benchmark datasets of live and spoof images from various capture technologies (standalone sensors, smartphone apps, etc.)
• Develop matching algorithms for noncontact compared to contact fingerprint images to address operational nonidealities in capture, such as blur and nonuniform illumination
• Evaluate match score differentials for various demographic groups (gender, ethnicity, finger size, etc.)
OUTCOMES FROM RESEARCH
• Database of non-contact finger photos and contact fingerprints
• Database of live and spoof non-contact finger photos
• Software for contact to non-contact finger matching
D.C. Keaton, A.S. Joshi, J.M. Dawson and N. Nasrabadi, “FDWST: Fingerphoto Deblurring using Wavelet Style Transfer,” IJCB 2024, Buffalo, NY, Sep 2024 (best student paper award)
A.S. Joshi, A. Dabouei, J.M. Dawson, and N. Nasrabadi. “UFQA: Utility guided Fingerphoto Quality Assessment,” IJCB 2024, Buffalo, NY, Sep. 2024
Hasan, et al 2022. Deep Coupled GAN-Based Score-Level Fusion for Multi-Finger Contact to Contactless Fingerprint Matching
Berti, A., et al 2022. Investigating the Impact of Demographic Factors on Contactless Fingerprints Interoperability
Jawade, B et al, 2022. RidgeBase: A Cross-Sensor MultiFinger Contactless Fingerprint Dataset
Purnapatra, S. et al, 2023. Presentation Attack Detection with Advanced CNN Models for Noncontact-based Fingerprint Systems
Hasan, M.M., et al, 2023. On Improving Interoperability for Cross-domain Multifinger Fingerprint Matching Using Coupled Adversarial Learning
Joshi, A.S., et al, 2023. Fingerphoto Deblurring Using Attention-Guided Multi-Stage GAN King, C., et al, 2023. Contactless Fingerprints: Differential Performance for Fingers of Varying Size and Ridge Density
Liveness Detection Competition- Noncontactbased Fingerprint Algorithms and Systems (LivDet-2023 Noncontact Fingerprint)
Adami, B. et. al., 2023. A Universal AntiSpoofing Approach for Contactless Fingerprint Biometric Systems
Noncontact finger photo
Contact fingerprint
Spoof noncontact finger photo
- CLARKSON UNIVERSITY
- UNIVERSITY AT BUFFALO
- WEST VIRGINIA UNIVERSITY
Face Morphing
With the advent of advanced generative AI, the vulnerability of merging of two faces into a single image has surfaced. The result is that two distinct individuals both match a single morphed image and are able to share an identity, which is potentially disruptive for digital identity systems. Given the diversity of approaches to create a morph, it’s difficult to develop adequate detectors to identify a morph. CITeR researchers are advancing from two sides. From the attackers side we are creating sophisticated databases of high quality morphs. From the protection side we are creating morph detection algorithms that detect a variety of morphs.
RESEARCH DIRECTIONS
• Face Morph Generation by Diffusion Models
• Creation of Face Morphs based on Generative Adversarial Networks
• Differential Morph Detection based on Transformers
OUTCOMES FROM RESEARCH
• Benchmark Datasets of High Quality Face Morphs
• Software for Differential Face Morph Detection
• Publications
N. Shukla and A. Ross, “Facial Demorphing via Identity Preserving Image Decomposition,” Proc. of International Joint Conference on Biometrics (IJCB), (Buffalo, USA), September 2024.
Blasingame, Z. and Liu, C., 2024. Leveraging diffusion for strong and high quality face morphing attacks. IEEE Transactions on Biometrics, Behavior, and Identity Science.
Blasingame, Z.W. and Liu, C., 2024. Greedy-DiM: Greedy Algorithms for Unreasonably Effective Face Morphs. arXiv preprint arXiv:2404.06025.
Neddo, R.E., Blasingame, Z.W. and Liu, C., 2024. The Impact of Print-and-Scan in Heterogeneous Morph Evaluation Scenarios. arXiv preprint arXiv:2404.06559.
Blasingame, Z., et al., 2023. Leveraging Diffusion for Strong and High Quality Face Morphing Attacks
Aghdaie, P., et al., 2023. Attention Augmented Face Morph Detection
Kashiani, H., et al., 2023. Towards Generalizable Morph Attack Detection with Consistency Regularization
Kashiani, H., et al., 2022. October. Robust ensemble morph detection with domain generalization
Zhang, N., et al., 2022. Fusion-based Few-Shot Morphing Attack Detection and Fingerprinting
Blasingame, Z. and Liu, C., 2021, August. Leveraging adversarial learning for the detection of morphing attacks. In 2021 IEEE International Joint Conference on Biometrics (IJCB) (pp. 1-8). IEEE.
Identity 1
Face morph
Identity 2
Research Highlights
- UNIVERSITY AT BUFFALO
- CLARKSON UNIVERSITY
Lip-syncing deepfake of President Kennedy.
Original video frame of President Kennedy.
Detection of Deepfakes
In the era of Artificial Intelligence, Deepfake technology has become one of the major threats to privacy, creativity, and authenticity. Deepfakes have extended influence across multiple domains, involving the manipulation of texts, audio, videos, images, and political and creative content. With growing concerns regarding the threats of deepfakes, CITeR researchers are developing ways to detect and prevent manipulated content.
RESEARCH DIRECTIONS
• Identification of authentic voice from synthesized voice, e.g., text-tospeech and voice conversion
• Detection of lip-syncing deepfake videos
• Development of multimodal deepfake video detection
• Evaluation of protection of deepfake injection attacks, such as challenge response
OUTCOMES FROM RESEARCH
• Software for detection of deepfake audio, images, and videos
• Datasets of deepfake audio and videos
Hou, S., Ju, Y., Sun, C., Jia, S., Ke, L., Zhou, R., Nikolich, A. and Lyu, S., 2024. DeepFake-O-Meter v2. 0: An Open Platform for DeepFake Detection. arXiv preprint arXiv:2404.13146.
Sun, C., Jia, S., Hou, S. and Lyu, S., 2023. AI-synthesized voice detection using neural vocoder artifacts. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 904-912).
Guo, H., Wang, X. and Lyu, S., 2023, June. Detection of real-time deepfakes in video conferencing with active probing and corneal reflection. In ICASSP 2023-2023
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 1-5). IEEE. Sun, C., et al, 2023. Aisynthesized voice detection using neural vocoder artifacts
Muppalla, S., et al, 2023. Integrating audio-visual features for multimodal deepfake detection
Kanti Datta, S., et al, 2024. Exposing Lip-syncing Deepfakes from Mouth Inconsistencies
Yan, Z., et al, 2023. Deepfakebench: A comprehensive benchmark of deepfake detection
Sun, C., et al, 2023. Using Vocoder Artifacts For Audio Deepfakes Detection
Yang, S., et al, 2023. Improving cross-dataset deepfake detection with deep information decomposition
Zhang, C., et al, 2023. Contrastive Multi-Face Forensics: An End-to-end Bi-grained Contrastive Learning Approach for Multi-face Forgery Detection
Ju, Y., et al, 2023. Glff: Global and local feature fusion for ai-synthesized image detection
Guo, H., et al, 2023, June. Detection of real-time deepfakes in video conferencing with active probing and corneal reflection
Sun, P., et al, 2023. FakeTracer: proactively defending against face-swap DeepFakes via implanting traces in training
WEST VIRGINIA UNIVERSITY
Presentation Attack Detection
Presentation attacks are a prevalent security concern today, where impostors attempt to gain access to restricted resources using fake biometric data such as face, fingerprint, or iris images. To mitigate these attacks, various presentation attack detection (PAD) systems have been deployed, often leveraging deep learning models for their high detection accuracy. CITeR researchers are working on creation of new spoof attacks, development of new PAD methods, and evaluation of these methods through open competitions and new datasets for development purposes.
RESEARCH DIRECTIONS
• Evaluate state-of-the-art algorithms through competitions — LivDet
• Develop software and hardware approaches for liveness detection — software PAD, 3D finger vein based on photoacoustics
• Prepare new and novel methods spoofing-face masks, realistic skin-colored finger spoofs, blood infused finger spoofs
OUTCOMES FROM RESEARCH
• Shared datasets
• LivDet competitions for face, iris, and fingerprint
R. Sharma, R. Sony, A. Ross, “Investigating WeightPerturbed Deep Neural Networks With Application in Iris Presentation Attack Detection,” Proc. of IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW), (Waikoloa, USA), January 2024.
D. Jauhari, R. Sharma, C. Chen, N. Sepulveda, A. Ross, “Iris Presentation Attack: Assessing the Impact of Combining Vanadium Dioxide Films with Artificial Eyes,” Proc. of IEEE/ CVF Winter Conference on Applications of Computer Vision Workshops (WACVW), (Waikoloa, USA), January 2024. Igene, L., Hossain, A., Chowdhury, M.Z.U., Rezaie, H., Rollins, A., Dykes, J., Vijaykumar, R., Komaty, A., Marcel, S., Tapia, J.E. and Aravena, C., Face Liveness Detection Competition (LivDet-Face)-2024.
Huang, M., Cai, J., Jia, S., Lokhande, V.S. and Lyu, S., 2024. MultiEdits: Simultaneous Multi-Aspect Editing with Text-toImage Diffusion Models. arXiv preprint arXiv:2406.00985. Blanton, M. and Murphy, D., 2024. Privacy preserving biometric authentication for fingerprints and beyond. Cryptology ePrint Archive.
Jawade, B., Subramanya, S., Dabhade, A., Setlur, S. and Govindaraju, V., 2024, May. GestSpoof: Gesture Based Spatio-Temporal Representation Learning for Robust Fingerprint Presentation Attack Detection. In 2024 IEEE 18th International Conference on Automatic Face and Gesture Recognition (FG) (pp. 1-9). IEEE.
Purnapatra, S., Liveness Detection CompetitionNoncontact-based Fingerprint Algorithms and Systems (LivDet-2023 Noncontact Fingerprint).
Micheletto, M., et al, 2023. Review of the fingerprint liveness detection (livdet) competition series: from 2009 to 2021.
Tinsley, P., et al, 2023, September. Iris Liveness Detection Competition (LivDet-Iris)–The 2023 Edition.
S. Purnapatra et al., “Presentation Attack Detection with Advanced CNN Models for Noncontact-based Fingerprint Systems,” 2023
Adami B, A universal anti-spoofing approach for contactless fingerprint biometric systems, 2023.
Face mask spoof.
Finger mold made from dental material.
Live finger and spoofs made of wood glue, playdoh, latex.
Clarkson University
Richard Plesh, PhD in Electrical Engineering, 2023
AI Scientist, Identity Sciences Lab (IDSL) at DHS Maryland Test Facility
“During my time working on CiTeR projects, I was focused on mitigating skin tone bias in facial recognition, creating a synthetic fingerprint generator for secure data sharing, and unraveling semantically meaningful features in deep learning facial recognition models. These projects not only deepened my understanding of biometrics but also honed my problem-solving skills and connected me with some of the best researchers in the field. This experience was instrumental in my academic and professional growth, culminating in my current role as an artificial intelligence scientist.”
University at Buffalo
Deen Dayal Mohan, PhD in Computer Science and Engineering, 2022 Research Scientist, Yahoo Research
“I began working on CITeR projects in 2017. Throughout my journey at CITeR, I primarily focused on face recognition and multimodal fusion/ aggregation methods. Being associated with CITeR for more than 5 years and working on these industry-related projects has helped me gain valuable experience, facilitating my transition into my industrial role.”
Michigan
State University
Renu Sharma, PhD in Computer Science and Engineering, 2022 Applied Scientist II, Amazon
“I worked on three research topics during my PhD program at MSU that contributed to the CITeR portfolio. The first project was on iris presentation attack detection, where the goal was to detect iris spoofs. The second project was on cross-modal biometrics, where the goal was to match near-infrared iris images against RGB face images. The third project was on generating and detecting iris morphs. These projects allowed me to hone my research skills and work on practical biometric problems from a fundamental science perspective.”
West Virginia University
Moktari Mostafa, PhD in Electrical Engineering, 2023
Oak Ridge Institute for Science and Education (ORISE) Fellow, Division of Imaging, Diagnostics, and Software Reliability (DIDSR) U.S. FDA
“I supported CITeR-funded projects from 2019 to 2023 on biometric identification. I developed sophisticated deep learning algorithms for face recognition, implemented cross-spectral iris matching systems, face video super-resolution, face frontalization, and pose-invariant facial recognition system. Each of these projects was an opportunity to solve challenging, real-world problems in biometrics.”
Renu Sharma
Deen Dayal Mohan
Richard Plesh
Moktari Mostafa
CITeR Fall 2022, Approved Proposals
(Performance period: 1/1/2023-12/31/2023)
– A Perpetual Deep Face Recognition System, 22F-01W
Nasser M. Nasrabadi (WVU), Mahedi Hasan (WVU)
– A Study to Benchmark Smartphone Hardware and Software for High Quality Iris Data Collection, 22F-01C
Soumyabrata Dey (CU), Masudul Imtiaz (CU)
– Detecting Real-time DeepFakes with Active Forensics and Biometrics, 22F-01B
Siwei Lyu (UB), Srirangaraj Seltur (UB)
– Effect of Specific Data Variations on Forensic Speaker Recognition Results, 22F-05W
Jeremy Dawson (WVU), Nasser Nasrabadi (WVU)
– Explainable Face and Fingerprint Matching via Improved Localization, 22F-05C
Faraz Hussain (CU), Daqing Hou (CU), Rashik Shadman (CU), Sarwar Murshed (CU)
– Large scale synthetically Generated face datasets (LEGAL2), 22F-01i
Sebastien Marcel (Idiap)
– Large-Scale Semi-Supervised Learning for Engine Audio Abnormality Detection and Understanding, 22F-04B
Srirangaraj Setlur, Venu Govindaraju (UB)
– Quality-Aware Deep Multimodal Biometric Recognition Systems, 22F-08W
Nasser Nasrabadi (WVU), Jerremy Dawson (WVU)
DHS Special Projects:
– Towards the Creation of a Large Dataset of HighQuality Face Morphs - Phase II, 22F-01J
Nasrabadi (WVU), Dawson (WVU), Li (WVU), Liu (CU), Schuckers (CU), Doermann (UB), Setlur (UB), Lyu (UB)
– Fully Homomorphic Encryption in Biometrics: Phase 2, 22F-02J
Nalini Ratha (UB), Vishnu Boddeti, Arun Ross (MSU)
– Scenario Testing for Presentation Attack Detection: Test Design and Requirements for Government Applications, 22F-08C
Stephanie Schuckers (CU), Masudul Imtiaz (CU)
CITeR Spring 2023, Approved Proposals
(Performance period: 7/1/2023-6/30/2024)
– An Approach to Develop a Large Non-contact and Contact-based Fingerprint Dataset with GANs to Foster Research, 23S-01CW
S. Purnapatra (CU), S. Konain Abbas (CU), F. Hussain (CU), S. Dey (CU), S. Schuckers (CU), J. Dawson (WVU)
– Modified EKF-ResNet Architecture for Automatic Speaker Recognition from Noisy Environment, 23S-07C
Sandip Purnapatra (CU), Masudul Imtiaz (CU
– Performance Benchmark: Ear-only vs. Ear+Face Fusion Biometrics - Adult and Children, 23S-08CB
Afzal Hossain (CU), Masudul Imtiaz (CU), Nalini Ratha (UB)
– Performance Evaluation of Cross-Spectral Iris Matching: Visible vs NIR, 23S-05W
Jeremy Dawson (WVU)
– Privacy-Preserving Biometric Matching over Homomorphically Encrypted Features, 23S-01MB
Vishnu Boddeti and Arun Ross (MSU), Nalini Ratha (UB)
– Utilization of Raw Photoacoustic Signal for Biometric Identification after Stress Activity, 23S-06B
Jun Xia (UB), Srirangaraj Setlur (UB)
Special Projects:
– Unmasking Mobile Biometrics, 23S-09C-SP
Stephanie Schuckers (CU), Daqing Hou (CU)
Gateway:
– Deepfake Attacks on Biometric Recognition: Evaluation of Resistance to Presentation and Injection Attacks (Gateway) 23S-02C-G
Masudul Imtiaz (CU), Daqing Hou (CU), Stephanie Schuckers (CU)
UNC-Charlotte/Clarkson University
Dr. Stephanie Schuckers
Director 315-268-6536
sschucke@clarkson.edu
Clarkson University
Laura Holsopple
Managing Director
315-268-2134 lholsopp@clarkson.edu
Clarkson University
Dr. Daqing Hou
Site Co-Director
315-268-7675
dhou@clarkson.edu
West Virginia University
Dr. Matthew Valenti
Site Director
304-293-9139 matthew.valenti@mail.wvu.edu
West Virginia University
Jeremy Dawson
Associate Professor
304-293-4028
jeremy.dawson@mail.wvu.edu
University at Buffalo
Dr. Venu Govindaraju
Site Director
716-645-3321
venu@cubs.buffalo.edu
University at Buffalo
Srirangaraj Setlur
Site Co-Director
716-645-1568 setlur@buffalo.edu
Michigan State University
Dr. Arun Ross
Site Director
517-353-9731
rossarun@cse.msu.edu
Idiap Research Institute (international site)
Dr. Sébastien Marcel
Site Director
+41 27 721 77 27 marcel@idiap.ch
UNC Charlotte (planned)
Dr. Heather Lipford
Site Director
704-687-8376
heather.lipford@charlotte.edu