Dr. Aleksandra Pižurica is Professor in statistical image modelling at Ghent University (Dipl. Ing 1994; Magister of Science 1997, PhD 2002). Her research is in the area of signal processing and machine learning, including multiresolution statistical image models, Markov Random Field models, sparse coding, representation learning, image and video reconstruction, restoration and analysis.

Prof. Pižurica serves currently as a Senior Area Editor for the IEEE Transactions on Image Processing and an Associate Editor for the IEEE Transactions on Circuits and Systems for Video Technology. She was also an Associated Editor for the IEEE Transactions on Image Processing (2012-2016) and the lead guest editor for the EURASIP Journal on Advances in Signal Processing for the Special Issue "Advanced Statistical Tools for Enhanced Quality Digital Imaging with Realistic Capture Models" (2013). The work of her team has been awarded twice the Best Paper Award of the IEEE GRSS (Geoscience and Remote Sensing Society) Data Fusion contest, in 2013 and 2014. She received the scientific prize "de Boelpaepe" for 2013-2014, awarded by the Royal Academy of Science, Letters and Fine Arts of Belgium.

Art investigation - an emerging and rapidly growing cross-disciplinary field of research - relies increasingly on signal processing and machine learning. Multimodal imaging is now routinely employed in order to support the technical study of art work, its conservation, or even presentation. Analysis of multimodal data uncovers often regions or patterns of interest that would otherwise remain unnoticed and enables this way new insights and even support for certain decisions that are made during the conservation-restoration treatments. In this talk, we discuss recent advances in digital signal processing and machine learning for supporting the study and restoration of paintings. The focus will be on sparse coding, representation learning and deep learning, spatial context modelling and Bayesian inference in tasks such as crack detection, paint loss detection and virtual inpainting. Concrete examples will be shown from the ongoing conservation-restoration treatment of the Ghent Altarpiece.

Dr. Andrea Tonello is professor of embedded communication systems at the University of Klagenfurt, Austria. He has been associate professor at the University of Udine, Italy, technical manager at Bell Labs-Lucent Technologies, USA, and managing director of Bell Labs Italy where he was responsible for research on cellular technology. He is also the founder of the spin-off company WiTiKee. He received the PhD from the University of Padova, Italy, and several awards among which the Lucent Bell Labs Recognition of Excellence Award, the RAENG (UK) Distinguished Visiting Fellowship, the IEEE VTS and COMSOC Distinguished Lecturer Awards, the Italian Full Professor Habilitation. He also received nine best paper awards. He serves/ed as associate editor for IET Smart Grid, IEEE ACCESS, IEEE Trans. on Communications, AEÜ Int. Journal of Electronics and Communications, and for the IEEE Trans. on Vehicular Technology. He was the chair of the IEEE Technical Committee on PLC, the general chair of IEEE SmartGridComm 2014 and IEEE ISPLC 2011. His current research spans the areas of wireless and power line communications, statistical signal processing, energy grids, and aerial robotics.

The introduction of renewables and the goal of increasing efficiency and reliability is revolutionizing the energy grids. Energy grids must become intelligent and be able to manage energy flows and flexibly adapt to changes in production and demand. This paradigm has fostered the creation of new business models on top of technology advances. The physical infrastructure must embed sensing, communication and processing capabilities. This talk will provide an overview of relevant applications (from metering, to diagnostics, to energy management) and discuss the role of sensing, processing and inference. It will also cover relevant communication technologies to provide connectivity and in particular new applications of power line communication that enable grid diagnostics by processing high frequency signals. In this context, we will also elaborate on the concept of learning the system model and discuss how top-down phenomenological approaches followed by machine learning paradigms can be merged with bottom-up physical domain approaches. .

Gregor Rozinaj is a Professor at the Faculty of Electrical Engineering and Information Technologies, Slovak University of Technology in Bratislava, currently he is a director of the Institute for Multimedia ICT, FEI STU. He published approx. 150 papers in scientific journals and international conferences. He is author of 4 patents, 3 of them worldwide. His scientific school consists of 14 successfully finished PhD students, more than 120 MSc students. He has served as principal investigator on more than 20 research projects, among them Horizon2020/FP7 projects. His research interest is oriented to multimedia processing, optimization techniques, fast algorithms, HCI. Previously he worked at the University of Stuttgart, Germany and Alcatel Research Centre in Stuttgart, Germany. (h-index: 6, Google scholar: 9).

Educational process is as old as mankind itself. Education of young generation is a complex task and every suitable tool has been used to improve the process. Nowadays, not only modern approach like student centric model, gamification, personalization, experience based learning, problem based learning, project orientation rather than knowledge acquisition has been introduced, but new technologies like virtual reality (VR), augmented reality (AR), interactive multimedia services offer better experience-based learning, faster education and more durable knowledge. Technology solutions are deeply dependent on economic situation of schools. One of the open problems of on-line education is distant presence of students and teachers. Telereality is a new keyword and covers research area of full virtual access to distant place with 3D audio and video stream and in specific case with multisensorial experience. Many projects have been devoted to the online education. Authors have many years of experience in online education within dozens of mostly European projects. The results of these projects show that online education is an extremely complex task with multidisciplinary problems. However, the projects show promising results in the topic. Results of the new Horizon 2020 NEWTON will be discussed, too.

Prof. Dr Milan Z. Bjelica is an associate professor at the University of Novi Sad, and Head of Innovation at RT-RK Research Institute. During his career he led the implementation of numerous solutions for the industry in the fields of consumer electronics, Internet of Things, multimedia and automotive, which are utilized by several major international Tier 1 and OEM companies. Some of the latest contributions include his work on central car computer design, providing heterogeneous environments for efficient execution of automotive algorithms, as well as providing means for deep learning execution in a transparent, controllable and safe environment. Dr Bjelica authored over a hundred scientific and professional publications, out of which 11 in major journals, as well as 24 patents and numerous technical contributions. He won "Special Merit Award" for his talk at the Consumer Electronics Show - ICCE in Las Vegas, 2013. He was nominated as the best young researcher by the Faculty of technical sciences in 2014. He is an active member of IEEE, especially IEEE Consumer Electronics Society within which he was a technical program chair for IEEE ICCE conferences in Berlin, Germany (2015, 2017) and one of the founders of IEEE ZINC conferences in Novi Sad, where he also chaired the program (2016-2018). He is also a regular visiting researcher and speaker at major events, including Fraunhofer Institute (Germany) and Consumer Electronics Association (USA).

Deep Learning is a promising field, allowing increase in Artificial Intelligence applications across many fields, ranging from data science, medical, weather, aerospace and automotive. Applications of Computer Vision - based deep learning are vastly assisted by modern System on Chip architectures, which provide the required parallelism, heterogeneity and interfacing. However, application of deep learning for safety-critical contexts where human lives might be at stake, such as in self driving cars, still has many pitfalls. Ongoing academic research tackles transparent AI, in which the correctness of AI is attempted to be reached by design; however, the outcome of this research is still far fetched. In this talk we will discuss a practical approach when integrating deep learning vision-based solutions in a safety-critical context, which can be achieved today. We outline approach which introduces a software/hardware platform design which fosters diversity, with the goal to minimize risk of critical failures which are induced by AI in decision making.

Robert Cupec graduated at the Faculty of Electrical Engineer-ing, University of Zagreb in 1995, where he received the M.Sc. degree in 1999. Upon graduation he joined the Department of Control and Computer Engineering in Automation, University of Zagreb, where he was employed until 2000. From 2000 to 2004 he is working at the Institute of Automatic Control Engineering, Technische Universität München, where he received the Ph.D. degree in 2005. Currently, he is employed as a Full Professor at the Faculty of Electrical Engineering, Computer Science and Information Technologies Osijek, J. J. Strossmayer University of Osijek, Croatia. His main research interest is in robot vision.

The ability to perceive a scene as a set of objects and recognize objects of interest is very important for intelligents robots which are expected to operate in complex and unstructured environments. Hence, this topic is of great interest to both the computer vision and robotics research community. In this tutorial, approaches for segmenting 3D point clouds acquired by a 3D kamera into objects, recognition of precisely defined objects of interest and classification of objects into one of previously learned object classes are considered. Plenty of methods solving the considered tasks are proposed. In this tutorial, we focus on methods suitable for application on service robots in households and public institutions as well as for automated agriculture. For these applications, besides the detection of the presence of an object of interest or identification of its class, the accurate object pose with respect to the robot as well as identification of object parts is required, since this information is often needed for planning a robot manipulation operation.