Title: Visualizing Feature Extraction and Classifier Design: An Application to Face Recognition
Speaker: Dr. Jamuna Kanta Sing
Professor, Department of Computer Science and Engineering
Jadavpur University, Kolkata 700032, India
Email: email@example.com, firstname.lastname@example.org
Abstract: Biometric technology has been used for recognizing a person based on physiological and/or behavioural characteristic such as fingerprint, palmprint, face, handgeometry, iris, signature, etc. However, among them face is extensively used due to its easy availability and can also be obtained from an unsupportive person. Although, face recognition from still images has matured over the couple of decades, the growing number of security breaches day-by-day has compelled researchers to study in-depth in this field, especially when dealing with surveillance videos. The need for highly secured and reliable face recognition technologies are becoming apparent in government offices, banking sectors, law enforcement, health and social services. In this tutorial, different types of feature extraction techniques with a focus on holistic approaches and design of suitable classifier based on neural networks will be presented in detail for development of reliable technology. Among the feature extraction techniques, the principal component analysis (PCA), two-dimensional PCA (2DPCA), Fisher’s linear discriminant analysis (LDA or FLDA), 2DLDA and generalized 2DFLDA (G-2DFLDA) will be presented in detail so as to visualizing the development of more advanced techniques. Whereas, in classifier design, implementation of radial basis function neural networks (RBFNN) will be discussed in detail.
Brief Biography: Jamuna Kanta Sing has received his B.E. (Computer Science & Engineering) degree from Jadavpur University in 1992, M.Tech. (Computer & Information Technology) degree from Indian Institute of Technology (IIT) Kharagpur in 1994 and Ph.D. (Engineering) degree from Jadavpur University in 2006. Dr. Sing has joined the Department of Computer Science & Engineering, Jadavpur University in March 1997 and presently serving as a Professor since 2010. He is a recipient of the BOYSCAST Fellow of the Department of Science & Technology, Govt. of India for doing advanced research at the University of Pennsylvania and the University of Iowa, USA in 2006 and the UGC Research Award in 2014. He is a senior member of the IEEE, USA. He has published more than 40 research papers in reputed refereed International Journals and more than 60 papers in international conferences. He has supervised 11 PhD scholars and handled as principal investigator (PI) (including on-going) 5 R&D projects from the AICTE, UGC and DST of worth around ₹65 Lakhs. His research interest includes face recognition and detection, video analytics, medical image processing, computational intelligence and pattern recognition.
Title: Introduction and Applications of Deep Learning
Speaker: Prof. Shekhar Verma
Professor, Information Technology
Indian Institute of Information Technology Allahabad, India
Abstract: The Deep Learning tutorial is planned to be slightly theoretical in nature. The plan is to take a participant with little background of Machine Learning and seamlessly usher him/her into the exciting world of Deep Learning. The entire spectrum of deep learning shall be developed ab initio from CNN, RNN, Variational Autoencoders to GANs to provide a unified view of the development of the area.
The tutorial is intended for undergraduate/graduate students and young faculty planning to working in the area. People from industry and academia looking for exposure will also find the tutorial beneficial. There are no strict prerequisites but some background will greatly enhance the take away from the tutorial.
Brief Biography: Prof Shekhar Verma has received his BTech, MTech and PhD from IIT BHU, Varanasi. He is currently working as professor in Information Technology at Indian Institute of Information Technology Allahabad. He has published more than 100 research papers in reputed refereed International Journals and more than 80 papers in international conferences. He has supervised 20 PhD scholars and handled many R&D projects. He is a member of the “Machine Learning and Optimization Group” at IIIT Allahabad. His research interests include dimensionality reduction, Manifold regularization, Privacy Preserving Machine Learning and Deep Learning techniques.