Most authentication is done as a one-time procedure, before access to a
protected resource is granted to the authorized user. For some
situations, this one-time authentication may not be sufficient. A
highly secure application may wish to guard against 'session
hijacking', in which an unauthorized user forcibly takes over an
authenticated login session after the legitimate user has successfully
logged in. Continuous Authentication is thus necessary. A secure
system must be able to, at any time, determine whether or not the authorized
user is still the one using the system. Any change in the continued
presence of the user, whether by his temporary absence, or by an
imposter, must be quickly detected to avoid compromising security.
Continuous Authentication presents numerous challenges: (i) it must be done passively, since it would be impractical to keep interrupting the user to re-authenticate using traditional methods; (ii) it must have minimal computational overhead, as otherwise the system may be too slow to be usable; (iii) it must achieve low false accepts and false rejects; (iv) it must provide authentication certainty even when the user is not using the keyboard or mouse or other input devices.
This lecture will discuss these issues in depth, and explore techniques in which practical Continuous Authentication may be achieved. In particular, the choice of Soft Biometrics (defined as biometrics that, under normal circumstances, does not uniquely identify any particular person, eg. gender) instead of Hard Biometrics, may prove sufficient and efficient for Continuous Authentication. The ubiquity of smart mobile devices provides a new platform for Continuous Authentication to become more mainstream.
Dr. Terence Sim is an Associate Professor at the School of Computing,
National University of Singapore. His research lies primarily in these
areas: Face recognition and analysis, Biometrics, and Computational
Photography. He also actively engages in Computer Vision problems,
leveraging methods from Computer Graphics and Machine Learning. He
has published over a hundred research articles in various journals and
conferences, and has served as guest speakers at local and
international scientific conventions.
Dr. Sim has taught many courses, at both the undergraduate and graduate levels, including Introductory Programming, Discrete Mathematics, Multimedia Analysis, Digital Visual Effects, and Biometrics. He has won several teaching awards and has consistently achieved high student ratings, even for large classes of four hundred students.
From 2014 to 2016, Dr. Sim was President of the Pattern Recognition and Machine Intelligence Association, Singapore, an affliate of the International Association of Pattern Recognition. From 2006 to 2014, he also served as Chairman of Workgroup 6 of the ISO/IEC JTC 1/SC 37 Biometrics in Singapore. Dr. Sim won the Temasek Young Investigator Award in 2006.
Dr. Sim obtained his PhD from Carnegie Mellon University in 2002, his MSc from Stanford University in 1991, and his SB from the Massachusetts Institute of Technology in 1990. He is also a proud alumnus of Raffles Institution, the oldest school in Singapore.