Prof. Peter Eisert
Humboldt Universty Berlin, Germany
Peter Eisert is Professor for Visual Computing at the Humboldt University Berlin and heading the Vision & Imaging Technologies Department of the Fraunhofer Institute for Telecommunications - Heinrich Hertz Institute Berlin, Germany. He received the Dipl.-Ing. degree in Electrical Engineering "with highest honors" from the Technical University of Karlsruhe, Germany, in 1995 and the Dr.-Ing. degree "with highest honors" from the University of Erlangen-Nuremberg, Germany, in 2000. In 2001, he worked as a postdoctoral fellow at the Stanford University, USA, on 3D image analysis and synthesis as well as facial animation and computer graphics. In 2002, he joined the Image Processing Department at FhG-HHI, where he is coordinating and initiating numerous national and international 3rd party funded research projects with a total budget of more than 10 Mio Euros. He has published more than 150 conference and journal papers on the subject of 3D reconstruction, facial expression analysis and synthesis and image-based rendering. In 2002, he received the SPIE VCIP Young Investigator Award, in 2008 the best paper award of the CVPR Nordia Workshop, and in 2011 and 2012 the best poster awards of Eurographics. He is Associate Editor of the International Journal of Image and Video Processing and in the Editorial Board of the Journal of Visual Communication and Image Representation. His research interests include 3D image analysis and synthesis, face processing, image-based rendering, computer vision, computer graphics, as well as image and video processing.
Speech Title: 3D Image Analysis for Virtual Reality and Digital Storytelling
Abstract: The current progress in Virtual and Augmented Reality, especially with the development and availability of VR/AR glasses, has triggered the demand for high-quality 3D assets for the creation of virtual environments. In this talk, 3D image and video analysis methods will be presented that target the digitalization of real environments. This includes novel methods to capture omnidirectional scenes as well as dynamic objects and people. Since video-based representations usually lead to high realism but are difficult to adjust to properties of the virtual environment, a combination with semantic computer graphics objects will be shown. Such hybrid representations allow for pose and appearance modifications while preserving high visual quality. The methods are investigated for the user to create novels forms of digital storytelling.
Prof. Godfried T. Toussaint
New York University Abu Dhabi, United Arab Emirates
Godfried Toussaint is a Professor and the Head of the Computer Science Program at New York University Abu Dhabi, in Abu Dhabi, The United Arab Emirates. He is also an affiliate researcher in the Computer Science and Artificial Intelligence Laboratory at the Massachusetts Institute of Technology in Cambridge, MA, USA. For many years he taught and did research in the School of Computer Science at McGill University in Montreal, in the areas of information theory, pattern recognition, textile-pattern analysis and design, computational geometry, machine learning, music information retrieval, and computational music theory. In 2005 he became a researcher in the Centre for Interdisciplinary Research in Music Media and Technology, in the Schulich School of Music at McGill University. He is a founder and co-founder of several annual international conferences and workshops, including the ACM Symposium on Computational Geometry, and the Canadian Conference on Computational Geometry. He is an editor of several journals, including Computational Geometry: Theory and Applications, the International Journal of Computational Geometry and Applications, ISRN Geometry, and the Journal of Mathematics and the Arts. He received several distinguished awards including a Killam Fellowship from the Canada Council for the Arts, and in 2009 a Radcliffe Fellowship from Harvard University, where he spent one year at the Radcliffe Institute for Advanced Study, and one year in the Music Department. His research on the phylogenetic analysis of musical rhythms has been reported in several media, and was the focus of two Canadian television programs.
Speech Title: Is Complex the Opposite of Simple?
Abstract: The notions of simple and complex as they are defined and utilized in the sciences, arts, and philosophy are briefly reviewed. The relation between complexity and complete randomness is examined. It is argued that simplicity is not the opposite of complexity. To support this conclusion this talk presents the results of recent experiments concerned with modeling human perception of the complexity of visual two-dimensional binary images or patterns, using objective mathematical measures of the stimuli. Two types of mathematical measures are compared: measures of structural complexity based on the presence of mirror (reflection) symmetries, and measures that approximate the Kolmogorov complexity via upper bounds of varying degrees of tightness. The Kolmogorov complexity of an image is the length of the shortest possible description of the image. The upper bounds on the Kolmogorov complexity that are compared consist of the lengths of the shortest descriptions of the images in a specific hierarchy of description languages defined by Papentin.
Prof. Vit Vozenilek
Palacky University, Czech Republic
Prof. Vit Vozenilek is a full professor and Ph.D. supervisor in Faculty
of Science, Palacky University Olomouc, Czech Republic.
He got his master degree in geography and mathematics in 1988 in Masaryk University, Brno, Czech Republic. He got a Ph.D. degree in physical geography in 1992 in the same university. Then he worked at Dept. of Geography at Palacky University Olomouc, Czech Republic, as an assistant for cartography and spatial modeling for more than seven years. He has involved many key projects as a supervisor for geocomputation and scientific visualization. In 1993, he received research scholarship at Birbeck College, University of London, UK. In 2001 he established Dept. of Geoinformatics at Faculty of Science, Palacky University Olomouc, Czech Republic. He is still its head. In 20010-2014, he held the position of vice-president of Palacky University Olomouc.
His research interests include geovisualisation (map making, atlas compilation, cognition aspects of map use) and spatial modeling. In recently 10 years, Prof. Vozenilek has published 7 books, 5 atlases, 35 conference papers and 28 journal papers.
Prof. Vozenilek serves as vice-president of International Cartographic Association, vice-president of Czech Cartographic Society, a member of many scientific journals, boards, councils, and committees. He serves as a reviewer for many research funds and journals, including once with a high impact factor. Prof. Vozenilek gave many invited lectures at universities abroad – UK, USA, Israel, Germany, China, Poland etc. He was the keynote speakers in many international conferences.
Speech Title: Proper use of colour schemes for image data visualization
Abstract: With the development of information and communication technologies, new technologies lead to an exponential increase in the volume and types of data available. At this time of the information society, data provides one of the most important arguments for policy making, crisis management, research and education, and many other fields. An essential task for experts calls for sharing high-quality data providing the right information at the right time. Designing of data presentation can largely influence the user perception and the cognitive aspects of data interpretation. Significant amounts of data can be visualised in some way. One image can thus replace a considerable number of numeric tables and texts. The paper focuses on the accurate visualisation of data from the point of view of used colour schemes. Bad choose of colours can easily confuse the user and lead to the data misinterpretation. On the contrary, correctly created visualisations can make information transfer much simpler and more efficient.
Prof. Patrick Wang, Northeastern University, USA
Prof. Patrick S.P. Wang, PhD. Fellow, IAPR, ISIBM, WASE and IEEE and ISIBM Outstanding Achievement Awardee, and is Tenured Full Professor, Northeastern University, USA, iCORE (Informatics Circle of Research Excellence) Visiting Professor, University of Calgary, Canada, Otto-Von-Guericke Distinguished Guest Professor, Magdeburg University, Germany, Zijiang Visiting Chair, ECNU, Shanghai, China, as well as honorary advisory professor of several key universities in China, including Sichuan University, Xiamen University, East China Normal University, Shanghai, and Guangxi Normal University, Guilin.
Prof. Wang received his BSEE from National Chiao Tung University (Jiaotong University), MSEE from National Taiwan University, MSICS from Georgia Institute of Technology, and PhD, Computer Science from Oregon State University.
Dr. Wang has published over 26 books, 200 technical papers, 3 USA/European Patents, in PR/AI/TV/Cybernetics/Imaging, and is currently founding Editor-in-Chief of IJPRAI (International Journal of Pattern Recognition and Artificial Intelligence) , and Book Series of MPAI, WSP. In addition to his technical interests, Dr. Wang also published a prose book, “Harvard Meditation Melody” 《哈佛冥想曲》and many articles and poems regarding Du Fu and Li Bai’s poems, Beethoven, Brahms, Mozart and Tchaikovsky’s symphonies, and Bizet, Verdi, Puccini and Rossini’s operas.
Speech Title: IPR, Big Data and Applications
--- Security, Safer Transportation and Greener World in Interacrtive Learning Environment
Abstract: This talk is concerned with fundamental aspects of Intelligent Pattern Recognition (IPR) and applications. It basically includes the following: Basic Concept of Automata, Grammars, Trees, Graphs and Languages. Ambiguity and its Importance, Brief Overview of Artificial Intelligence (AI), Brief Overview of Pattern Recognition (PR), What is Intelligent Pattern Recognition (IPR)? Interactive Pattern Recognition Concept, Importance of Measurement and Ambiguity, How it works, Modeling and Simulation, Basic Principles and Applications to Computer Vision, Security, Road Sign Design, Safer Traffic and Robot Driving with Vision, Ambiguous (Dangerous and Bad) design of Road Signs vs Unambiguous (Good) Road Signs, How to Disambiguate an Ambiguous Road Sign? What is Big Data? and more Examples and Applications of Learning and Greener World using Computer Vision. Finally, some future research directions are discussed.
Prof. Junyu Dong
Ocean University of China, China
Prof. Junyu Dong received his BSc and MSc from the Department of Applied Mathematics at Ocean University of China in 1993 and 1999 respectively. From 2000 to 2003, he studied in the Department of Computer Science at Heriot-Watt University in the UK, and received his PhD in Image Processing in November 2003. Prof. Dong joined Ocean University of China in 2004. He is currently a professor and the head of the Department of Computer Science and Technology. His research interests include computer vision, underwater image processing and machine learning, with more than 10 research projects supported by NSFC, MOST and other funding agencies. He has published more than 100 journal and conference papers.
Speech Title: Perception Driven Texture Generation
Abstract: This talk introduces a novel task of generating texture images from perceptual descriptions. Perceptual attributes, such as directionality, regularity, roughness together with semantic descriptions are important factors for human observers to describe a texture. Although in the past procedural models and even commercial packages were commonly used for creating textures, texture generation from user-defined perceptual attributes or semantic descriptions has been rarely studied. In this paper, we propose a framework based on joint deep neural networks that combines adversarial training and perceptual feature regression for texture generation, while only random noise and user-defined perceptual attributes are required as input. In this framework, a preliminary trained convolutionalneural network is essentially integrated with the adversarial framework, which can drive the generated textures to possess given perceptual attributes. An important aspect of the proposed model is that, if we change one of the input perceptual features, the corresponding appearance of the generated textures will also be changed. We designed several experiments to validate the effectiveness of the proposed method. The results show that the proposed method can produce high quality texture images with desired perceptual properties. Demonstrations based on mobile applications and virtual reality are also provided.
Prof. Jianhao Tan
Hunan University, China
I am Jianhao Tan, aged 53, born in Liuyang, Hunan, China. I received the B.S. degree in the Forging Equipment and Engineering from
Huazhong Technology College, Wuhan, China, in 1983, and the M.S. degree in the Pressure Process from Huazhong University of
Engineering and Technology, Wuhan, China, in 1989, and the Ph.D . degree in Control Science and Engineering from Hunan University,
Changsha, China, in 2010.
From 1989 to 1997,I worked with Hunan Province Computing Technology Research Institute as an assistant-researcher. I has worked with
Hunan University since 1998.I has been a professor at Hunan University since 2008. My research interests include data mining, pattern
recognition, system identification, and image processing.
I published more than 30 pieces of articles, has undertaken more than 10 items of projects.
Speech Title: Smoothing Processing of Artificial Potential Field Method to the Preliminary Path Based on A* Algorithm
Abstract: The path planning problem in the 3D mountain environment is a multi-target optical one. In these targets, path length is conflicting with smoothness. An efficient fusion algorithm for the rotary-wing flying robot is presented for solving the above problem. This fusion algorithm combines A* algorithm with artificial potential field method both improved and optimized. Firstly, A* algorithm, which is optimized in the heuristic function, is adopted to plan the preliminary path with the shortest length but without good smoothness in the environmental model established by grid method. Then, the key nodes in the preliminary path are selected and saved. Finally, artificial potential field algorithm is used to do the path smoothing processing on the basis of these key nodes. It is testified from theory that the fusion algorithm can provide the most smoothing path. The simulations in MATLAB of the proposed algorithm, single A* algorithm and artificial potential field method are respectively carried out in the 3D map consisting of random irregular surface and defined peak. The simulation results indicates that the fusion algorithm is feasible in 3D path planning and superior in smoothness performance , which can satisfy the property of real-time tracking for rotary-wing flying robots.
Assoc. Prof. Dr. Matsumoto Mitsuharu,
University of electro-communications, Japan
Mitsuharu Matsumoto is currently an associate professor in the University of Electro-Communications. He received a B.E. in Applied Physics, and M.E. and Dr. Eng. in Pure and Applied Physics from Waseda University, Tokyo, Japan, in 2001, 2003, and 2006, respectively. His research interests include acoustical signal processing, image processing, pattern recognition, self-assembly, human-robot interaction and robotics. He received Ericsson Young Scientist Award from Nippon Ericsson K.K, Japan and FOST Kumada Award, in 2009 and 2011, respectively. He published around a hundred of journal and international conference papers. He is a member of the Institute of Electrical and Electronic Engineers (IEEE).
Speech topic: Nonlinear filters and its application to image processing
Abstract: In this speech, I pick up a nonlinear filter called epsilon-filter, and introduce its application to image processing. Epsilon-filter is a simple nonlinear filter developed about 30 years ago. Original filter is developed for noise reduction from the image, but several improved versions are developed for many applications. They are also applicable not only to image signal but also to audio signal as some features used in epsilon-filter are common in image and audio signals. Therefore, in this speech, I also introduce some examples of the application of epsilon-filter to audio signal and discuss their common points.
Assoc. Prof. Linhua Deng,
Yunnan Observatories, Chinese Academy of Science, China
Linhua Deng is currently an associate professor in Yunnan Observatories, Chinese Academy of Sciences. He received a Ph. D degree of astrophysics in 2014 in University of Chinese Academy of Sciences. His research interests include solar image processing, time series analysis, and pattern recognition, with more than ten research projects supported by NSFC, CAS, and other funding agencies. He has published more than 50 international journal and conference articles.
Speech topic: Hilbert-Huang Transform Analysis of Long-term Solar Magnetic Activity
Abstract: The quasi-periodic analysis of solar magnetic activity has been carried out by various authors during the past fifty years. In this talk, the novel Hilbert-Huang transform approach is applied to investigate the yearly numbers of polar faculae in the time interval from 1705 to 1999. The detected periodicities can be allocated to three components: the first one is the short-term variations with periods smaller than 11 years, the second one is the mid- term variations with classical periods from 11 years to 50 years, and the last one is the long-term variations with periods larger than 50 years. The analysis results improve our knowledge on the quasi-periodic variations of solar magnetic activity and could be provided valuable constraints for solar dynamo theory..