Narendra Ahuja

Professor Emeritus, UIUC, USA

Abstract: TBA

Rama Chellapa

Distinguished University Professor and Minta Martin Professor of Electrical and Computer Engineering, University of Maryland. Fellow IEEE, ACM, AAAI, AAAS

Title: Unconstrained Face Recognition using Deep Networks

Abstract: TBA

Venu Govindaraju

Distinguished Professor Department of Computer Science and Engineering, VP Research, State University of New York Buffalo. Fellow IEEE, ACM, IAPR, AAAS, SPIE

Title: Cognitive Biometrics

Abstract:Given the pervasive use of online transactions and personal data storage in the cloud, there is an urgency to develop a mechanism of robust authentication that protects the privacy and online assets of individuals. The biometrics and cyber security communities have approached the challenge of warding off intruders from different vantage points. The former focuses on "liveness" of biometric signals whereas the latter has primarily considered encryption and elaborate software protocols. This talk will explore holistic approaches that go beyond the traditional biometric modalities of physical and behavioral modalities by integrating the tests for humanness and identity in a cognitive framework. We will show how under the continuous authentication scenario requirements, our approach allows for a more practical approach to security.

Nalini Ratha

IBM T. J. Watson Research Center, USA

Title: Adversarial Attacks on Biometrics and Deep Learning Systems

Abstract:Biometrics systems are vulnerable to different types of attacks at various the processing susbtages.
Many attacks can be prevented by suitably designing the system components. In the first part of this talk, we will describe the attacks and present possible solutions that can thwart these attacks. Recently, deep learning systems have shown significant improvement in accuracy performance in biometrics systems. We will illustrate how such high accuracy system can be fooled by carefully designing inputs to create both false accept and false rejects in the system by carefully choosing the alterations in the input. In this part of this talk, we will explore possible solutions to this new problem while enjoying high accuracies of these systems.

Vishal Patel

Assistant Professor, Rutgers University, USA

Title: Continuous User Authentication on Mobile Devices

Abstract: Recent developments in sensing and communication technologies have led to an explosion in the use of mobile devices such as smartphones and tablets. With the increase in use of mobile devices, one has to constantly worry about the security and privacy, as the loss of a mobile device would compromise personal information of the user. To deal with this problem, continuous authentication (also known as active authentication) systems have been proposed in which users are continuously monitored after the initial access to the mobile device. This talk will provide an overview of different continuous authentication methods on mobile devices. We will discuss merits and drawbacks of available approaches and identify promising avenues of research in this rapidly evolving field.

Ajay Kumar

Hong Kong Polytechnical University

Abstract: TBA

Angshul Majumdar

Assistant Professor, IIIT-Delhi

Abstract: TBA

Saket Anand

Assistant Professor, IIIT-Delhi

Title: Distance Metric Learning

Abstract:Computing distances is central to many techniques designed for learning problems like classification, clustering and retrieval. These techniques frequently rely on default distance metrics such as the Euclidean or Cosine distance, which are often inconsistent with our semantic notion of (dis)similarity, resulting in poor performance at the learning task. Distance Metric Learning is a popular approach that circumvents this limitation by learning a task-specific distance function from the data itself. Over the last decade, Distance Metric Learning has come a long way - from learning a linear Mahalanobis distance function to learning complex neural network based nonlinear distance metrics. In this talk we will review some of the popular metric learning approaches and discuss some recent advances in this area. We will also explore different applications including face verification and person re-identification.

Richa Singh

Associate Professor, IIIT-Delhi

Abstract: TBA

Mayank Vatsa

Associate Professor, IIIT-Delhi

Abstract: TBA