Keynote Lectures

Human Behaviour Understanding for Automotive and Surveillance

Rita Cucchiara University of Modena and Reggio Emilia Italy

Brief Bio
Rita Cucchiara is Professor of Computer Architecture and Computer Vision at Engineering Department “Enzo Ferrari” of the University of Modena and Reggio Emilia, Italy.
She heads the Imagelab research lab at UNIMORE. She is Director of the Inter-departmental Centre of Research in ICT Softech-ICTof Emilai rOmagna High Technology network, and Director of the Master in Visual Computing and Multimedia Technology.
She is working on computer vision, multimedia pattern recognition, machine learning and intelligent sensing. In 2016 she has been awarded by the Facebook Artificial intelligence Research and the Cineca Italian SuperComputing Resource Allocation grants. Rita Cucchiara is president of GIRPR the Italian Association of Pattern Recognition, Fellow of IAPR and she is in the Advisory Board of Computer Vision Foundation. She covers also the role of deputy-president of the Italian Groups of Computer Engineering Professors GII, for scientific research. In 2020 she will be General Chair of ICPR2020. She is currently AE of IEEE Trans. of Multimedia. She is author or more than 300 publications on International Journals and Proceedings.

Abstract
Human behavior understanding (HBU) is a central topic for many different disciplines, from sociology, psychology to, more recently computer science. In this latter framework, Computer Vision and Pattern Recognition are strategic: advancements in motion analysis and interpretation, 2D and 3D video analysis and, obviously, deep learning make computer vision research in HBU a true success story.
The talk will focus on HBU for two very related contexts. The first is automotive, a key application area nowadays, due to the fact that cameras are considered mandatory sensing components in cars to support assisted or automatic guidance, to improve the safety and comfort of drivers and passengers. HBU is needed inside and outside the car: to understand what the driver is doing, what he/she can do to drive bettre and what people around the car are doing. The second is surveillance where the study of human presence and activity has a long tradition. Here the analysis of behaviour of pedestrian, groups of people and crowd is based on computer vision: it now achieves unbelievable results due to the presence of annotated datasets and deep solutions. Surveillance and automotive are becoming always more connected in smart cities environments.
In the talk, I will present an overview of recent research centred on humans in video acquired by the city or the car point of view. The talk will discuss the milestones of this research, the improvements and changes in effort due to deep machine learning approaches and the available annotated big data. Then I will present our research project at Imagelab on persons under surveillance systems and on persons in Human-Vehicle-Interaction applications. These solutions are mostly based on 3D information, saliency analysis and video segmentation and ground on extensive use of deep networks technology. I will discuss deep architectures we propose for understanding person presence, person behaviour and person interaction.

Time Evolving Networks in Finance and Medicine

Edwin Hancock* York University United Kingdom *Distinguished IAPR Speaker

Brief Bio
Edwin R. Hancock holds a BSc degree in physics (1977), a PhD degree in high-energy physics (1981) and a D.Sc. degree (2008) from the University of Durham, and a doctorate Honoris Causa from the University of Alicante in 2015. From 1981-1991 he worked as a researcher in the fields of high-energy nuclear physics and pattern recognition at the Rutherford-Appleton Laboratory (now the Central Research Laboratory of the Research Councils). During this period, he worked on high energy physics experiments at the Stanford Linear Accelarator Center (SLAC) providing the first measurements of charmed particle lifetimes. He also held adjunct teaching posts at the University of Surrey and the Open University. In 1991, he moved to the University of York as a lecturer in the Department of Computer Science, where he has held a chair in Computer Vision since 1998. He leads a group of some 25 faculty, research staff, and PhD students working in the areas of computer vision and pattern recognition. His main research interests are in the use of optimization and probabilistic methods for high and intermediate level vision. He is also interested in the methodology of structural and statistical and pattern recognition. He is currently working on graph matching, shape-from-X, image databases, and statistical learning theory. His work has found applications in areas such as radar terrain analysis, seismic section analysis, remote sensing, and medical imaging. He has published about 170 journal papers and 610 refereed conference publications. He was awarded the Pattern Recognition Society medal in 1991 and an outstanding paper award in 1997 by the journal Pattern Recognition. He has also received best paper prizes at CAIP 2001, ACCV 2002, ICPR 2006, BMVC 2007 and ICIAP in 2009 and 2015. In 2009 he was awarded a Royal Society Wolfson Research Merit Award. In 1998, he became a fellow of the International Association for Pattern Recognition. He is also a fellow of the Institute of Physics, the Institute of Engineering and Technology, and the British Computer Society. In 2016 he became a fellow of the IEEE and was named Distinguished Fellow by the British Machine Vision Association. He is currently Editor-in-Chief of the journal Pattern Recognition, and was founding Editor-in-Chief of IET Computer Vision from 2006 until 2012. He has also been a member of the editorial boards of the journals IEEE Transactions on Pattern Analysis and Machine Intelligence, Pattern Recognition, Computer Vision and Image Understanding, Image and Vision Computing, and the International Journal of Complex Networks. He has been Conference Chair for BMVC in 1994 and Progrmme Chair in 2016, Track Chair for ICPR in 2004 and 2016 and Area Chair at ECCV 2006 and CVPR in 2008 and 2014, and in 1997 established the EMMCVPR workshop series. He has been a Governing Board Member of the IAPR since 2006, and is currently Vice President of the Association.

Abstract
This talk focusses on how to use network entropy as a means of characterising network structure and investigating the relationship between changes in network structure and function with time. Examples are presented on network data extracted from the data for the New York Stock Exchange. We show how the entropic characterisation can be extended to develop Euler- Lagrange equations which describe the evolution of the node degree distribution, and can be used to predict the evolution of network structure with time. If time permits, we will also describe how to extend our model to include quantum spin statistics, and explore how Bose-Einstein and Fermi-Dirac statistics modify the evolution of network structure. We demonstrate some of the utility of the proposed methods on fMRI images of Alzheimer brains.

Biomedical Imaging - Challenges and Potentials

Xiaoyi Jiang University of Münster Germany

Brief Bio
Xiaoyi Jiang studied Computer Science at Peking University, China, and received his PhD and Venia Docendi (Habilitation) degree from University of Bern, Switzerland. He was associate professor at Technical University of Berlin and since 2002 full Professor at University of Münster, Germany. Currently, he is the dean of Faculty of Mathematics and Computer Science at University of Münster. He is a PI and research area leader of the Cluster of Excellence “Cells in Motion – Imaging to understand cellular behaviour in organisms” established by the German Research Foundation DFG in 2013. He is Editor-in-Chief of International Journal of Pattern Recognition and Artificial Intelligence. In addition, he also serves on the Advisory Board and Editorial Board of several journals, including IEEE Transactions on Medical Imaging, International Journal of Neural Systems, Pattern Analysis and Applications, and Pattern Recognition. His research interests include biomedical imaging, 3D image analysis, and structural pattern recognition. He is a Senior Member of IEEE and Fellow of IAPR.

Abstract
Imaging has become an indispensable tool in biology and medicine for both basic research and clinical practice. The specific image characteristics and problems in these fields have motivated researchers to develop novel concepts and algorithms. This talk emphasizes the fundamental research view of biomedical imaging and discusses a number of related challenges, concepts, and algorithms. In addition to the traditional computer vision approaches, another focus will be given to machine learning based approaches. In particular, Barista (an open-source graphical high-level interface for the Caffe deep learning framework) and its application to biomedical imaging will be presented. Besides the information processing view the imposing development in biomedical imaging also provides a driving force for life sciences from a Galisonian perspective.

On Location and Registration Fiducials: Their Analysis and Design

Alfred Bruckstein Technion Israel

Brief Bio
Alfred M. Bruckstein, born in Transylvania, Romania, in 1954, received his BSc and MSc degrees at the Technion, Haifa, in 1976 and 1980, respectively and then earned a Ph.D. degree in Electrical Engineering in Stanford University, California in 1984, his advisor being Professor Thomas Kailath. From October 1984 he has been with the Technion, where he presently holds of the Ollendorff Chair in Science, in the Computer Science Department. His research interests are in Ants and Swarm Robotics, Signal and Image Processing, Image Analysis and Synthesis, Pattern Recognition, and various aspects of Applied Geometry. Professor Bruckstein authored and co-authored over one hundred and fifty journal papers in the fields of interest mentioned. From 2002 till 2005 he served as the Dean of The Technion Graduate School, and from 2006-2011 as the Head of Technion’s Excellence Program for Undergraduate Studies. Since 2009, he is also affiliated with the Nanyang Technological University in Singapore, as a Visiting Professor in the Department of Mathematics. Professor Bruckstein is a member of the AMS, and MAA, and a SIAM Fellow for contributions to Signal Processing, Image Analysis, and Ant Robotics, and received SIAM’s 2014 SIAG-Imaging Science Prize (with David Donoho and Michael Elad, for the paper “From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images”).

Abstract
A location/registration fiducial is a shape or pattern designed to provide, via a sensing or imaging device (usually a camera), information about the absolute or relative location of objects in space. This talk will discuss methods for the design of such fiducials, and ways to mathematically analyze and predict the localization performance obtained when sensing them.