Prof. Jean Christophe Pesquet
Title: Unfolding proximal algorithms
Abstract: We show in this talk how proximal algorithms, which constitute a powerful class of optimization methods, can be unfolded under the form of deep neural networks. This yields to improved performance and faster implementations while allowing to build more explainable, more robust, and more insightful neural network architectures. Application examples in the domain of image restoration will be provided.
Prof. Andrea CAVALLARO
Queen Mary University,
School of Elec. Eng and Computer Science,
Queen Mary University of London
Mile End Road, London E1 4NS,
Title: Adversarial attacks on image classifiers
Images of people and scenes we share online reveal information about personal choices and preferences, which can be automatically inferred by classifiers. To prevent privacy violations and to protect the visual content from unwanted automatic inferences, we show how to exploit the vulnerability of classifiers to adversarial attacks to craft adversarial perturbations that maintain (and even improve) image quality. However, these perturbations may be ineffective against classifiers that were not seen during the generation of the perturbation or against defences that use re-quantization or image compression. To address these limitations, I will present a series of adversarial attacks designed to protect visual content against classifiers. I will discuss how to craft perturbations based on randomised ensembles to make them robust to defences, on image semantics to selectively modify colours within chosen ranges that are perceived as natural by humans, and on perturbations that enhance image details.
Andrea Cavallaro is Professor of Multimedia Signal Processing and the founding Director of the Centre for Intelligent Sensing at Queen Mary University of London, UK. He is Fellow of the International Association for Pattern Recognition (IAPR) and Turing Fellow at the Alan Turing Institute, the UK National Institute for Data Science and Artificial Intelligence. He received his Ph.D. in Electrical Engineering from the Swiss Federal Institute of Technology (EPFL), Lausanne, in 2002. He was a Research Fellow with British Telecommunications (BT) in 2004/2005 and was awarded the Royal Academy of Engineering Teaching Prize in 2007; three student paper awards on target tracking and perceptually sensitive coding at IEEE ICASSP in 2005, 2007 and 2009; and the best paper award at IEEE AVSS 2009. Prof. Cavallaro is Editor-in-Chief of Signal Processing: Image Communication; Chair of the IEEE Image, Video, and Multidimensional Signal Processing Technical Committee; an IEEE Signal Processing Society Distinguished Lecturer; and an elected member of the IEEE Video Signal Processing and Communication Technical Committee. He is Senior Area Editor for the IEEE Transactions on Image Processing and Associate Editor for the IEEE Transactions on Circuits and Systems for Video Technology. He is a past Area Editor for the IEEE Signal Processing Magazine (2012-2014) and past Associate Editor for the IEEE Transactions on Image Processing (2011-2015), IEEE Transactions on Signal Processing (2009-2011), IEEE Transactions on Multimedia (2009-2010), IEEE Signal Processing Magazine (2008-2011) and IEEE Multimedia. He is a past elected member of the IEEE Multimedia Signal Processing Technical Committee and past chair of the Awards committee of the IEEE Signal Processing Society, Image, Video, and Multidimensional Signal Processing Technical Committee. Prof. Cavallaro has published over 270 journal and conference papers, one monograph on Video tracking (2011, Wiley) and three edited books: Multi-camera networks (2009, Elsevier); Analysis, retrieval and delivery of multimedia content (2012, Springer); and Intelligent multimedia surveillance (2013, Springer).
Prof. Eric GRANGER
Dept. of Systems Engineering,
École de technologie supérieure (Université du Québec),
1100, rue Notre-Dame Ouest Montréal (Qc), H3C 1K3,
Title: Video Recognition from Large and Weakly-Labeled Image Datasets
State-of-the-art sensors to capture audio-visual signals are paving the way to innovative, next-generation technologies for many important applications in, e.g., video surveillance, medical diagnosis, health monitoring, content-based image/video retrieval. For instance, the detection, tracking and recognition of actions, cars, people, etc., appearing over a distributed network of cameras is a key component for many video-based summarization and surveillance applications. Designing accurate recognition systems for these applications typically gives rise to several challenges because it involves learning complex models using large weakly-annotated data sets that incorporate domain shifts, subtle noise, variations and uncertainties embedded in real-world signals. Focused on training deep learning models from large and weakly-labeled image datasets, this talk will discuss recent advances in weakly-supervised learning, domain adaptation, and context-based fusion, which promise to address such complex video processing problems.
Eric Granger received Ph.D. in Electric Engineering from École Polytechnique de Montréal in 2001, and worked as a Defense Scientist at Defense R&D Canada (1999-2001), and in R&D with Mitel Networks (2001-04). He joined the École de technologie supérieure (Université du Québec), Montreal, in 2004, where he is presently Professor of Systems Engineering, and director of LIVIA, a research laboratory focused on computer vision and artificial intelligence. His research interests include machine learning, computer vision, pattern recognition, domain adaptation, and incremental and weakly-supervised learning, with applications in affective computing, biometrics, medical image analysis, and video analytics and surveillance.
Prof. Mouloud ADEL
Aix-Marseille Université, Institut Fresnel UMR-CNRS 7249,
Domaine universitaire de Saint-Jérôme,
Avenue escadrille Normandie Niemen,
13397 Marseille cedex 20,
Title: Computer-Aided Diagnosis on Medical Images : Application to neurodegenerative diseases classification tasks.
Computer-Aided Diagnosis (CAD), based on medical image analysis, is in the interface of medicine and computer science. It is designed to help doctors to make quantitative evaluation of several diseases. CAD systems are generally based on several steps: data acquisition, pre-processing, feature extraction and classification. Recently, deep learning techniques has given an impressive performance in recognition and classification tasks. Therefore, a great number of researches turned to use neural network to address CAD. In this tutorial i will focus on a CAD system on Positrons Emission Tomography (PET) brain images for neurodegenerative diseases for classification purposes and more specifically on Alzheimer’s Disease. Feature selection as well as classification tasks will be described and discussed to compare classical approaches to neural-network ones.
Mouloud Adel is a Professor in Computer Science and Electrical Engineering at Aix-Marseille University, Marseille, France. His research areas concern signal and image processing and machine learning applied to biomedical and industrial images. He has been involved in many international research programs and he is a member of the editorial board of Journal of Biomedical Engineering and Informatics. He has been an invited speaker at different universities and has been the co-organizer of various international conferences and workshops. He also served as a regular reviewer, associate and guest editor for a number of journals and conferences. He chaired a special session "Statistical Image Analysis for computer-aided detection and diagnosis on Medical and Biological Images" in IPTA 2014, Paris.
Prof. Oge MARQUES
Florida Atlantic University,
College of Engineering and Computer Science,
Florida Atlantic University (FAU),
Boca Raton, FL,
Title: Deep learning for medical imaging analysis (and beyond): latest advances and recipes for success
The use of deep learning solutions, architectures, tools, and techniques in the field of medical image analysis has experienced tremendous growth and expressive results in the fields of dermatology, ophthalmology, and multiple specialties of radiology in the past few years. In this tutorial we cover some of the latest developments in the field of medical image analysis using deep learning, explain why the development of algorithms and deep learning models for solving specific problems are just the beginning, and suggest a recipe for success in the field.
Oge Marques is a Professor of Computer Science and Engineering in the College of Engineering and Computer Science and, by courtesy, a Professor of Information Technology in the College of Business at Florida Atlantic University (FAU) (Boca Raton, FL). He received his PhD in Computer Engineering from FAU in 2001 and a Master's in Electronic Engineering from Philips International Institute (the Netherlands). He is a world-renowned expert in the area of intelligent processing of visual information, which encompasses the fields of image processing, computer vision, human vision, artificial intelligence (AI) and machine learning. His current research focuses on the intersection of AI and medicine. He is the author of ten technical books, one patent, and more than a hundred scientific articles in his fields of expertise.
Dr. Marques is a Fellow of the Leshner Leadership Institute of the American Association for the Advancement of Science (AAAS), an ACM (Association for Computing Machinery) Distinguished Speaker, and a Sigma Xi Distinguished Lecturer. He is also a Tau Beta Pi Eminent Engineer, a Senior Member of both the IEEE (Institute of Electrical and Electronics Engineers) and the ACM (Association for Computing Machinery), and a member of the honor societies of Sigma Xi, Phi Kappa Phi and Upsilon Pi Epsilon. Prof. Marques has more than 30 years of teaching experience in different countries (USA, Austria, Brazil, Netherlands, Spain, France, and India). He has won several teaching awards, including: the FAU College of Engineering and Computer Science Senior Faculty Teaching Award (2020), the Engineers' Council John J. Guarrera Engineering Educator of the Year
Award (2019), the Outstanding Mid-Career Teaching Award, American Society for Engineering Education - Southeastern Section (ASEE-SE) (2011), and the Excellence and Innovation in Undergraduate Teaching Award, FAU, three times (2018, 2011, 2004).
Dr. Liu HANTAO
5 The Parade, Cardiff, CF24 3AA,
Title: Image quality assessment and saliency modelling
Reliably predicting visual media quality as perceived by humans remains challenging and is of high practical relevance. A significant research trend is to investigate visual saliency and its implications for visual quality assessment. Fundamental problems regarding how to acquire reliable eye-tracking data and how saliency should be incorporated in computational quality assessment models are largely unsolved. This talk will focus on methodologies for reliably collecting eye-tracking data, assessment of the capabilities of saliency in improving the performance of quality assessment models, as well as the optimised use of saliency in visual quality assessment systems.
Dr Hantao Liu is Associate Professor with the School of Computer Science and Informatics, Cardiff University, United Kingdom. He received the Ph.D. degree from the Delft University of Technology, Delft, The Netherlands, in 2011. He is a founder member of the Delft Image Quality Lab. His research interests include visual media quality assessment, visual attention modelling and applications, visual scene understanding, medical image perception and human-machine interaction. He is serving as Associate Editor for the following international journals: IEEE Transactions on Multimedia; IEEE Transactions on Human-Machine Systems; Signal Processing: Image Communication.