Keynote Talks

Prof. Nasir Memon​
​New York University


Title: Emerging NUI-based Methods for User Authentication

Abstract: As user demand and cost benefits of natural user interface technologies are hastening their adoption, computing devices that come equipped with these natural interaction interfaces are becoming ubiquitous. Consequently, authentication mechanisms on them are becoming an essential security component to enable a wider range of applications that need higher requirements of security as well as privacy.
In this talk we will survey the lan​d​scape of ”point-of-entry” user-device authentication mechanisms based on behavioral biometrics that require a natural user interaction using gestural​ ​or non-gestural interaction for access. This interaction includes 2-D touch gestures, 3-D gestures, voice, eye tracking, and brain​ ​computer​ ​interaction. We will analyze their potential security and usability promises and issues, and discuss plausible solutions that could be pursued in future work.

Biography: Nasir Memon is a professor in the Department of Computer Science and Engineering at New York University (NYU) Tandon School of Engineering. He is one of the founding members of the Center for Cyber Security (CCS), a collaborative initiative of multiple schools within NYU including NYU- Steinhardt, NYU ‐ Wagner, NYU - Law. His research interests include digital forensics, biometrics, data compression, network security and security and human behavior.
Memon earned a Bachelor of Engineering in Chemical Engineering and a Master of Science in Mathematics from Birla Institute of Technology and Science (BITS) in Pilani, India. He received a Master of Science in Computer Science and a PhD in Computer Science from the University of Nebraska. He has published over 250 articles in journals and conference proceedings and holds a dozen patents in image compression and security. He has won several awards including the Jacobs Excellence in Education award and several best paper awards. He has been on the editorial boards of several journals and was the Editor‐In‐Chief of Transactions on Information Security and Forensics. He is an IEEE Fellow, an SPIE Fellow and was a distinguished lecturer of the IEEE Signal Processing Society.

Dr. Gang Hua​​​
Microsoft Research Asia


Title: Towards Efficient and Accurate Subject Level Face Representation

Abstract: The effectiveness of a face recognition algorithm is determined by the quality of the face representation. Most previous methods construct the face representation at the instance level, i.e., each face image of a subject is represented separately. We argue that a compact, comprehensive, and discriminative subject-level representation that integrates information from all face image of the same subject is more desirable, which has the potential to support both more scalable and more accurate face recognition.
In this talk, I will present a Neural Aggregation Network (NAN) for an efficient and accurate subject level face representation. The network takes a set of variable number of face images of a person as its input, and produces a compact and fixed-dimension visual representation of that person. The whole network is composed of two modules. The feature embedding module is a CNN which maps each face frame into a feature representation. The neural aggregation module is composed of two content based attention blocks which is driven by a memory storing all the features extracted from the set of face images through the feature embedding module. The output of the first attention block adapts the second, whose output is adopted as the aggregated representation of the video faces. Due to the attention mechanism, this content-aware representation is invariant to the order of the face images. The experiments show that the proposed NAN consistently outperforms natural aggregation methods such as average pooling, and achieves state-of-the-art accuracy on three video face recognition datasets: the YouTube Face, IJB-A, and Celebrity-1000 datasets.
Towards the end of the talk, I will also discuss on a potential security hole of current deep network based face recognition system, which suggests that state-of-the-art computer-based face recognition system is still far away from achieving the level of robustness as human. We will also discuss on a principled method to counter such attacks.

Biography: Gang Hua is a Senior Research Manager in the Visual Computing Group at Microsoft Research Asia. He was an Associate Professor of Computer Science in Stevens Institute of Technology between 2011 and 2015. He held an Academic Advisor position at IBM T. J. Watson Research Center between 2011 and 2014. He was a visiting researcher at Microsoft Research Asia in Summer 2013, and a Consulting Researcher at Microsoft Research in Summer 2012. He had also worked as full-time Researchers at leading industrial research labs for IBM T. J. Watson Research Center, Nokia Research Center Hollywood, and Microsoft Live Labs Research. He received the Ph.D. degree in Electrical and Computer Engineering from Northwestern University in 2006.
His research in computer vision studies the interconnections and synergies among the visual data, the semantic and situated context, and the users in the expanded physical world, which can be categorized into three themes: human centered visual computing, big visual data analytics, and vision based cyber-physical systems. He is the author/coauthor of more than 130 peer reviewed publications in prestigious international journals and conferences. His research was funded by NSF, NIH, ARO, ONR, Adobe Research, Google Research, Microsoft Research, and NEC Labs.
Dr. Gang Hua founded the first Face Recognition SDK for Microsoft in 2008, which has evolved to the Face API in Microsoft Cognitive Service. He is the recipient of the 2015 IAPR Young Biometrics Investigator Award for his contributions to Unconstrained Face Recognition in Images and Videos. He is elected as an IAPR Fellow in the 2016 class for his contributions to Visual Computing and Learning from Unconstrained Images and Videos. He is also named as a 2016 ACM Distinguished Scientist for his contributions to Multimedia and Computer Vision. To date, he holds 18 U.S. patents and has more than 11 U.S. patents pending. He is a Senior Member of the IEEE.