||Hosted by |
There are four tutorials offered to the participants of ICB 2016. The registration details for these tutorials are available in the registration page.
Tutorial Morning 1: Privacy Preserving Biometric Identity Verification
Gérard Chollet, Dijana Petrovska-Delacrétaz, Bhiksha Raj
(June 13, 9:00-12:00)
More information: PDF
Biometric identify verification systems provide many advantages over conventional password-based authentication systems. Unlike the latter, which could be forgotten of pilfered, the former are based on indelible physical attributes, and are not dependent on potentially faulty memory.
On the other hand, biometric authentication systems come with privacy risks. The authenticating server now possesses an equally indelible biometric signature of the subject. The server, or any malicious adversary who manages to “steal” the signature from it, or even a legal entity who subpoenas the server, could abuse these signatures in a variety of ways, such as using them to identify the subject in other unauthorized contexts (e.g. searching for their faces of voices on YouTube videos, or matching fingerprints in potential crime scenes), creating fake signals (e.g. synthetic voices or fingerprints) that could then be used to authenticate as the user, etc.
The need, then, is for secure biometric authentication systems in which guarantees exist that the biometric signatures of users possessed by the system cannot be abused even by the system itself.
The obvious solution is to encrypt the biometric signatures at the server, such that the server itself (let alone any third-party entity who accesses the server) cannot decrypt it, and to perform the biometric matching at the server using encrypted signatures themselves. This is a task for “homomorpic encryption” – encryption techniques that enable computations such as matching in the encrypted domain. Homomorphic encryption techniques, however, remain practically infeasible, in spite of recent progress in the area.
The purpose of this tutorial is to review the current issues and describe alternative and complementary approaches to homomorphic encryption for secure biometric authentication. In this context we will briefly review several techniques including secure multiparty computation, oblivious transfer, zero-knowledge proofs, garbled circuits, locality sensitive hashing, and binary embeddings, and also discuss computational and performance tradeoffs. Cancellable and revocable biometrics will also be reviewed.
Gérard Chollet studied Linguistics, Electrical Engineering and Computer Science at the University of California, Santa Barbara where he was granted a PhD in Computer Science and Linguistics. He taught at Memphis State University and University of Florida before joining CNRS. In 1981, he was asked to take in charge the speech research group of Alcatel. In 1983, he joined a newly created CNRS research unit at ENST. The group contributed to a number of European projects such as SAM, ARS, FreeTel as well as national projects. In 1992, he was asked to participate to the development of IDIAP, a new research laboratory of the `Fondation Dalle Molle' in Martigny, Switzerland. IDIAP contributed to SpeechDat, M2VTS and other European projects. From 1996 to 2012, he was full time at ENST, managing research projects and supervising doctoral work. Funding was secured from such projects as Eureka-Majordome and MajorCall, NoE-BioSecure, Strep-SecurePhone, IP-Companion@ble, AAL-vAssist, FET-ILHAIRE,... He supervised more than forty doctoral thesis. CNRS decided in july 2012 to grant him an emeritus status. He visited Boise State University in 2013 and the University of Eastern Finland in 2014. He is now VP of Res. of Intelligent Voice (http://www.intelligentvoice.com/).
Dijana Petrovska-Delacrétaz obtained her degree in Physics and her PhD from the Swiss Federal Institute of Technology (EPFL) in Lausanne. She was working as a Consultant at AT&T, as a post-Doc at Télécom ParisTech, and as a Senior Scientist in the Informatics Department of Fribourg University, Switzerland. Since 2004 she is an associate professor in Mines Télécom / Télécom SudParis. Her research activities are oriented towards pattern recognition, signal processing, and data-driven machine learning methods, that are exploited for applications such as speech, speaker and language recognition, very low-bit speech compression, biometrics (2D and 3D face, and voice), and privacy preserving biometrics (cancelable biometric and generation of cryptographic keys from biometric data). Her publication list is composed of three patents, two publicly available databases (for speaker recognition and biometrics evaluations), open-source software for reproducible results, 81 publications, and co-supervision of five PhD thesis.
Bhiksha Raj is an Associate Professor in the Language Technologies Institute of the School of Computer Science at Carnegie Mellon University, with additional affiliations to the Electrical and Computer Engineering and Machine Learning departments. Dr. Raj obtained his PhD from CMU in 2000 and was at Mistubishi Electric Research Laboratories from 2001-2008. Dr. Raj's chief research interests lie in automatic speech recognition, computer audition, machine learning and data privacy. Dr. Raj's latest research interests lie in the newly emerging field of privacy-preserving speech processing, in which his research group has made several contributions.
Tutorial Morning 2: 3D morphable face model and its applications
Josef Kittler, Paul Koppen
(June 13, 9:00-12:00)
More information: PDF
3D Morphable Face Models (3DMM) have been used in face recognition for some time now. They can be applied in their own right as a basis for 3D face recognition and analysis involving 3D face data. However their prevalent use over the last decade has been as a versatile tool in 2D face recognition to normalise pose, illumination and expression of 2D face images. It has the generative capacity to augment the training and test databases for various 2D face processing related tasks. It can expand the gallery set for pose invariant face matching. For any 2D face image it can furnish complementary information, in terms of its 3D face shape and texture. It can also aid multiple frame fusion by providing the means of registering a set of 2D images. A key enabling technology for this versatility is 3D face model to 2D face image fitting. The recent developments in 3D model to 2D image fitting will be discussed. They include the use of symmetry to improve the accuracy of illumination estimation, multistage close form fitting to accelerate the fitting process, modifying the imaging model to cope with 2D images of low resolution, and building albedo 3DMM. These various enhancements will be overviewed and their merit demonstrated on a number of face analysis related problems.
Josef Kittler is professor of Machine Intelligence at the Centre for Vision, Speech and Signal Processing, University of Surrey. He received his BA, PhD and DSc degrees from the University of Cambridge. He teaches and conducts research in the subject area of Signal Processing and Machine Intelligence, with a focus on face biometrics, and anomaly detection. He published a Prentice Hall textbook on Pattern Recognition: A Statistical Approach and several edited volumes, as well as more than 700 scientific papers, including in excess of 180 journal papers. He serves on the Editorial Board of several scientific journals in Pattern Recognition and Computer Vision. He became Series Editor of Springer Lecture Notes on Computer Science in 2004. He served as President of the International Association for Pattern Recognition 1994-1996.
Paul Koppen, PhD, is the Project Manager for a major collaborative research project in face recognition FACER2VM, funded by the Engineering and Physical Sciences Research Council. He is based at the University of Surrey, where he studied for a PhD degree in 3D face analysis. After his PhD he was employed as a postdoctoral research fellow working on a project relating face shape to genetics, funded by the Wellcome foundation. More recently he had responsibility for technology transfer in the area of 3D face model building and 3D face morphable model fitting.
Tutorial Afternoon 1: Contactless 3D Fingerprint Acquisition and Matching
(June 13, 13:30-16:30)
More information: PDF
Contact-based 2D fingerprint identification is widely employed for the civilian and law-enforcement applications around the world. Such traditional acquisition of fingerprint images by pressing or rolling of fingers against the hard surface (glass, silicon, polymer) or paper often results in partial or degraded images due to improper finger placement, skin deformation, slippages, smearing or due to sensor noise. As a result full potential from the fingerprints is not realized. Therefore, touchless 3D finger imaging systems have emerged to address above intrinsic problems. Such 3D approaches can also provide more accurate personal identification as rich information is available from 3D fingerprint images.
Emerging solutions for the contactless 3D fingerprint acquisition are largely based on shape from silhouette, structured lighting or photometric stereo based imaging. However widely accepted standards or the representation of 3D fingerprint features is yet to emerge. The minutiae features are widely considered to be most reliable and widely employed by law enforcement experts and commercial 2D fingerprint systems available today. Accurate recovery, representation, selection, registration and matching of 3D fingerprints is essentially a biometrics recovery/alignment, and matching problem. This half-day tutorial will provide algorithmic details relating to 3D fingerprint recovery, matching and interoperability of 3D fingerprints.
Ajay Kumar received the Ph.D. degree from the University of Hong Kong, Hong Kong, in 2001. He was an Assistant Professor with the Department of Electrical Engineering, IIT Delhi, Delhi, India, from 2005 to 2007. He is currently working as Associate Professor in the Department of Computing, Hong Kong Polytechnic University, Hong Kong. His current research interests are on biometrics with an emphasis on hand biometrics, vascular biometrics, iris, and multimodal biometrics. He holds several U.S. patents, and has authored extensively on biometrics and computer vision-based industrial inspection. He is an area editor for the Pattern Recognition Letters Journal and served on the IEEE Biometrics Council as the Vice President (Publications) during 2011-2015. He was on the Editorial Board of the IEEE Transactions on Information Forensics & Security from 2010 to 2013, and served on the program committees of several international conferences and workshops in the field of his research interest. He was the Program Chair of the Third International Conference on Ethics and Policy of Biometrics and International Data Sharing in 2010, the Program Co-Chair of the International Joint Conference on Biometrics held in Washington, DC, in 2011, the International Conference on Biometrics held in Madrid, in 2013, CVPR 2013- 2016 Biometrics Workshops, and has also served as General Co-Chair for the Second International Joint Conference on Biometrics in 2014.
Tutorial Afternoon 2: BEAT with hands-on: an online web-platform for reproducible research in computational science
Andre Anjos, Laurent El Shafey, Sebastien Marcel
(June 13, 13:30-16:30)
More information: PDF
This tutorial will present the BEAT platform for online reproducible research, introducing concepts and providing an initial hands-on experience. The BEAT platform allows novice and advanced researchers to: (1) benchmark systems and components; (2) run comparative evaluations; (3) attest (certify) toolchains; (4) provide educational material for new-comers in pattern recognition and (5) optimize algorithms and systems. All these tasks can be accomplished without installing additional software on the users computer, running exclusively from the web browser. The BEAT platform naturally enforces important research aspects such as reproducibility and component re-use.
Andre Anjos (http://andreanjos.org) received his Ph.D. degree in signal processing from the Federal University of Rio de Janeiro in 2006. He joined the ATLAS Experiment at European Centre for Particle Physics (CERN, Switzerland) from 2001 until 2010 where he worked in the development and deployment of the Trigger and Data Acquisition systems that are nowadays powering the discovery of the Higgs boson. During his time at CERN, Andr´e studied the application of neural networks and statistical methods for particle recognition at the trigger level and developed several software components still in use today. In 2010, Andr´e joined the Biometrics Group at the Idiap Research Institute where he works mostly with face biometrics. His current interests include reproducible research in biometrics, anti-spoofing and recognition using faces, pattern recognition, image processing and machine learning. Andre currently leads the design and implementation of the BEAT platform for evaluation and testing. He also serves as reviewer for several scientific journals in pattern recognition, image processing and biometrics.
Laurent El Shafey received his Ph.D. in Electrical Engineering in 2014 from Ecole Polytechnique Federale de Lausanne (EPFL), Switzerland. He holds a Master in Computer Science from the TU Darmstadt, Germany and a Master in Electrical Engineering from Supelec, France. Laurent was a post-doctoral researcher in the Biometric Person Recognition Group at Idiap Research Institute, Switzerland, when he contributed to the development of the BEAT platform. He now works at Google. His research interest is in machine learning and biometrics with a focus on face and speaker recognition. He is the recipient of the EAB Biometrics Research Awards 2014.
Sebastien Marcel (http://www.idiap.ch/~marcel) received the Ph.D. degree in signal processing from Universit´e de Rennes I in France (2000) at CNET, the research center of France Telecom (now Orange Labs). He is currently interested in pattern recognition and machine learning with a focus on biometrics. He is a senior researcher at the Idiap Research Institute (Switzerland), where he heads a research team and conducts research on face recognition, speaker recognition, vein recognition and spoofing attacks detection. In 2010, he was appointed Visiting Associate Professor at the University of Cagliari (IT) where he taught a series of lectures in face recognition. He is also lecturer at the Ecole Polytechnique F´ed´erale de Lausanne (EPFL) where he is teaching on “Fundamentals in Statistical Pattern Recognition”. He serves on the Program Committee of several scientific journals and international conferences in pattern recognition and computer vision. He is Associate Editor of IEEE Transaction on Information Forensics and Security (TIFS) since 2013. He is also co-Editor of the upcoming “Handbook of Biometric Anti-Spoofing” with Prof M. Nixon and Prof. S.Z. Li that will be published by Springer. Finally, he is Guest Editor of an IEEE TIFS Special Issue on “Biometric Spoofing and Countermeasures”. Sebastien Marcel is the principal investigator of international research projects including MOBIO (EU FP7 Mobile Biometry - https://themobioproject.org/), TABULA RASA (EU FP7 Trusted Biometrics under Spoofing Attacks - http://www.tabularasa-euproject.org) and BEAT (EU FP7 Biometrics Evaluation and Testing - http://www.beat-eu.org). Finally he is also the Director of the Swiss Center for Biometrics Research and Testing (http://www. biometrics-center.ch).
|Center for Applied Intelligent Systems Research (IS-Lab/CAISR), School of Information Technology, Halmstad University, Sweden|