Biometrics, Spring 2020

Jan 14, 2020

Details

Course: CSE 40537 / 60537 Biometrics
Level: Undergraduate and Graduate
Instructor: Daniel Moreira (dhenriq1@nd.edu)

~~Lectures: TUE and THR, 5:05 to 6:20 PM, 125 DeBartolo Hall~~¹
~~Office Hours: MON and WED, 2:00 to 4:00 PM, 150N Fitzpatrick Hall~~¹
Lectures: TUE and THR, 2:00 to 3:15 PM, at Zoom¹
Office Hours: TUE and THR, 5:05 to 6:20 PM, at Zoom¹
Students are not obligated to attend classes at 2:00 pm, but are certainly welcome. All classes are being recorded with Panopto.

Slack: ~~https://cse-biometrics-spr20.slack.com~~ (now deactivated)
Panopto: https://bit.ly/33ZkU97
Zoom: https://notredame.zoom.us/my/dmoreira

Course grades are now available.

Progress

01/14/2020 - Syllabus, Course details.
01/16/2020 - Basics I, Biometrics, traits, and systems.
01/21/2020 - Basics II, Errors, metrics, and attacks.
01/23/2020 - 1st Coding Class, Implementation of metrics.
01/28/2020 - Fingerprint Recog. I, History and features.
01/30/2020 - Fingerprint Recog. II, Acquisition and enhancement.
02/04/2020 - Fingerprint Recog. III, Minutiae detection.
02/06/2020 - Fingerprint Data Collection, password with instructor.
02/11/2020 - 2nd Coding Class, Minutiae-based recognition.
02/13/2020 - 2nd Coding Class, continuation.
02/18/2020 - Face Recog. I, Why faces and faces vs. fingerprints.
02/20/2020 - Face Recog. II, Acquisition and enhancement.
02/25/2020 - Face Recog. III, Description and matching.
02/27/2020 - 3rd Coding Class, Face recognition.
03/03/2020 - Fingerprints assignment, Developers’ day.
03/05/2020 - Fingerprints assignment, Attackers’ day.
03/10/2020 - Spring Break.
03/12/2020 - Spring Break.
03/17/2020 - Extended Spring Break¹.
03/18/2020 - Extended Spring Break¹.
03/24/2020 - New Course Directions¹.
03/26/2020 - Iris Recog. I, Why irises and irises vs. other traits.
03/31/2020 - Iris Recog. II, Acquisition and enhancement.
04/02/2020 - Iris Recog. III, Description and matching.
04/07/2020 - 4th Coding Class, Iris recognition.
04/09/2020 - Multibiometrics, Other traits, data fusion.
04/14/2020 - 1st Invited Talk, Dr. Andrey Kuehlkamp.
04/16/2020 - 2nd Invited Talk, Dr. Adam Czajka.
04/21/2020 - Faces assignment, Developers’ day.
04/23/2020 - Irises assignment, Developers’ day.
04/28/2020 - Assignment Report due date.
05/04/2020 - Final exam, see grades.

Important Dates

~~03/03/2020 - Fingerprints assignment, Developers’ day.~~
~~03/05/2020 - Fingerprints assignment, Attackers’ day.~~
~~03/31/2020 - Faces assignment, Developers’ day.~~¹
~~04/02/2020 - Faces assignment, Attackers’ day.~~¹
~~04/14/2020 - Irises assignment, Developers’ day.~~¹
~~04/16/2020 - Irises assignment, Attackers’ day.~~¹
~~04/28/2020 - Last assignment, Collaboration day.~~¹
~~04/14/2020 (5:05 PM at Zoom) - Dr. Andrey Kuehlkamp’s talk.~~
~~04/16/2020 (5:05 PM at Zoom) - Dr. Adam Czajka’s talk.~~
~~04/21/2020 - Faces assignment, Developers’ day¹.~~
~~04/23/2020 - Irises assignment, Developers’ day¹.~~
~~04/28/2020 - Final Report due date¹.~~
~~05/04/2020 - Final exam.~~

Invited Talks


	Dr. Andrey Kuehlkamp Postdoctoral Research Associate at the Center for Research Computing, University of Notre Dame


Diverse Aspects in Advancing Iris Recognition Systems Are we ready for widespread, mass-scale adoption of iris recognition systems? Following the miniaturization of fingerprint scanners, these have dominated recognition systems and have even become almost commonplace for unlocking cell phones, but what if in the not-so-far-off future they were replaced with iris scanners, would you be comfortable with it? Since its initial introduction in 1993, automated iris recognition has dramatically grown in popularity and soon could become the dominant method for automated recognition. Take for example the largest recognition system in the world — India’s Aadhaar program — which has collected more than 1.1 billion irises from their citizens to be used as the primary identification for banking, pensions, and welfare programs. Even more recently — November 2017 — Somaliland became the first country in the world to use iris recognition as the means for identification in a public election, which had more than 800,000 registered voters. Although a mature technology in many regards, the drastic increase in iris recognition adoption has revealed many opportunities for improvement. In this talk I present an overview of my research, which focuses on improving iris recognition in three ways: speed, accuracy, and robustness.

Diverse Aspects in Advancing Iris Recognition Systems

Are we ready for widespread, mass-scale adoption of iris recognition systems? Following the miniaturization of fingerprint scanners, these have dominated recognition systems and have even become almost commonplace for unlocking cell phones, but what if in the not-so-far-off future they were replaced with iris scanners, would you be comfortable with it? Since its initial introduction in 1993, automated iris recognition has dramatically grown in popularity and soon could become the dominant method for automated recognition. Take for example the largest recognition system in the world — India’s Aadhaar program — which has collected more than 1.1 billion irises from their citizens to be used as the primary identification for banking, pensions, and welfare programs. Even more recently — November 2017 — Somaliland became the first country in the world to use iris recognition as the means for identification in a public election, which had more than 800,000 registered voters. Although a mature technology in many regards, the drastic increase in iris recognition adoption has revealed many opportunities for improvement. In this talk I present an overview of my research, which focuses on improving iris recognition in three ways: speed, accuracy, and robustness.


	Dr. Adam Czajka Assistant Professor at the Department of Computer Science and Engineering, University of Notre Dame


Is this eye alive or artificial? Oh wait, maybe it’s dead? Detection of unknown presentation attacks in biometrics. Presentation attacks are those physical presentations to a biometric system that aim at driving it into an incorrect decision. Rediscovered recently in general computer vision community (and raising a significant interest; look — for instance — for famous stop sign attacks on deep learning-based object detection models), these attacks are known in biometrics for several decades. In this talk, I will use iris recognition as an example and will present the huge creativity of attackers in using various artifacts (printouts, patterned contact lenses, plastic eyes, GAN-generated fakes and… dead eyes) to spoof a system. Although training a model to recognize each of these presentation attack instruments is relatively easy and works well, a big challenge now is how to build models that generalize onto unknown attack types, going beyond our understanding of attackers’ creativity when training our models. I will present a few methods we are exploring in our research to provide presentation attack detection methods that perform promisingly in open-set classification scenario.

Is this eye alive or artificial?
Oh wait, maybe it’s dead? Detection of unknown presentation attacks in biometrics.

Presentation attacks are those physical presentations to a biometric system that aim at driving it into an incorrect decision. Rediscovered recently in general computer vision community (and raising a significant interest; look — for instance — for famous stop sign attacks on deep learning-based object detection models), these attacks are known in biometrics for several decades. In this talk, I will use iris recognition as an example and will present the huge creativity of attackers in using various artifacts (printouts, patterned contact lenses, plastic eyes, GAN-generated fakes and… dead eyes) to spoof a system. Although training a model to recognize each of these presentation attack instruments is relatively easy and works well, a big challenge now is how to build models that generalize onto unknown attack types, going beyond our understanding of attackers’ creativity when training our models. I will present a few methods we are exploring in our research to provide presentation attack detection methods that perform promisingly in open-set classification scenario.

Biometrics on the News

Posted by the students the instructor on Slack:

COVID-19

1: Modified/canceled due to COVID-19.

Acknowledgments

This course is heavily based on Dr. Adam Czajka’s and Dr. Walter Scheirer’s previous Biometrics courses. I sincerely thank them for kindly allowing me to rely upon their materials.

Daniel Moreira

Assistant Professor of Computer Science

Computer scientist with interests in (but not limited to) Computer Vision, Machine Learning, Media Forensics, and Biometrics.