iV2013 - courses

This half-day course provides a hands-on introduction for Augmented Reality technology and its application to visualisation and art. It is aimed at the beginner’s level, it requires no previous knowledge and programming skills. All a participant needs is a laptop and an iPhone, iPad or Android phone.

The pre-conference short course is structured as follows:

1. Introduction

2. A theoretical introduction to AR

3. Example projects and demos

4. Hands-on tutorials

5. Marker-based AR using Layar

6. Location-based AR using Layar and Hoppala

7. Animated 2.5D AR objects using Layar

8. 3D AR objects using Junaio

9. Creative individual mini-projects

10. Summary and discussion

The course is self-contained and is intended to provide its participants with both theoretical understanding and practical skills they need to start their own AR projects immediately.

Biography

Dr Vladimir Geroimenko is Professor of Multimedia and Web Technology at Plymouth University, UK and also a researcher and digital artist, specializing in Augmented Reality. He has taught AR technology to undergraduate and postgraduate students since 2005. His AR art projects can be found at his online Art Gallery – www.geroimenko.com.

Vladimir is the editor of the pioneering book “Augmented Reality Art: From an Emerging Technology to a Novel Creative Medium”, written by a team of world-leading artists, researchers and practitioners and to be published by Springer Verlag in 2014. He is the organizer of a symposium on Augmented Reality Visualisation and Art at this Conference.

vgeroimenko@plymouth.ac.uk

www.plymouth.ac.uk/staff/vgeroimenko

A Full-day Course: Monday 15 July 2013, Time: 10:00 - 17:00

Art for Visualizers

Francis T. Marchese, Pace University, NewYork, USA

http://csis.pace.edu/~marchese

Abstract

The confluence of art and visualization has a long history. Indeed, the Paleolithic artists who painted on the cave walls of southwest France may have been the first visualizers. Or was it vice versa? Either way, throughout the intervening millennia visual artists have become proficient at transforming information into representations that are designed to communicate and provoke. The challenge facing a viewer of art is how to decipher an image’s content and extract its meaning. This holds true for a viewer of visualizations as well.

Thus, the purpose of this tutorial is to introduce the fundamental skills for analyzing visual art that subsequently may be applied to scientific and information visualizations. It will offer an historical survey of the intersections of art and visualization with an emphasis on examples from contemporary artists, and provide an opportunity for participants to practice these skills within a gallery setting. To this end, the tutorial will be composed of two sessions. A morning session will focus on an historical survey, conceptual foundations, and skill acquisition. An afternoon session convening at one of London’s art museums will allow course participants to test their analysis skills on a selection of gallery’s paintings.

Level of Tutorial: Introductory

Biography

Frank Marchese has a Ph.D. in quantum chemistry from the University of Cincinnati and was a National Institutes of Health Postdoctoral Research Fellow specializing in the statistical mechanics of liquids. He is currently Professor of Computer Science at Pace University where he teaches courses in computer graphics, data visualization, human-computer interaction, and software engineering. His research interests span scientific and information visualization; novel user interfaces for visualization; distributed and collaborative visualization; integration of visualization into lifecycles for scientific research and software engineering; and the development of visualization systems at the intersection of art, science, and technology.

He is founder and Director of Pace’s Center for Advanced Media (CAM) and the Pace Digital Gallery, the latter of which is a collaboration between Pace University’s Seidenberg School of Computer Science and Information Systems and Department of Fine Arts. He has published widely in science, technology, and art; and is editor of the conference proceedings entitled Understanding Images published by Springer-Verlag.

Frank has been twice awarded Pace’s School of Computer Science and Information Systems Excellence in Research Award, received the Kenan Award for Teaching Excellence, and been nominated for The Carnegie Foundation Teacher of the Year Award. In December 2008, he was awarded Pace University’s Faculty Award for Distinguished Service. Most recently he has been a visiting scholar at New York University’s Institute of Fine Arts where he has extended his scholarship into museum curatorial studies, installation of art in alternative spaces, and the relationship between text and image in medieval art. He is currently exploring the artistic origins of information visualization.

A Full-day Course: Monday 15 July 2013, Time: 10:00 - 17:00

Information Visualization – A course

Bob Spence, Imperial College London, UK

ABSTRACT

Bob Spence is the author of one of the two (equally) most popular textbooks on information visualization. His one-day course is directed, not at researchers, but rather to two groups of people. One group comprises students who come to information visualization for the first time: they can come from any discipline, especially since no knowledge of computer science or mathematics is required. The other group potentially interested in the course comprises those who have to teach the subject and who wish to see one approach to that task.

Biography

Bob Spence has been conducting research into Human-computer Interaction, and information visualization in particular, since 1968. He regularly presents an updated course on information visualization every year at Imperial College London, the Technical University of Eindhoven in The Netherlands and, from 2013, Madeira University (Portugal). Bob is a Fellow of the Royal Academy of Engineering.

A half-day Course: Monday 15 July 2013, Time: 13:30 - 17:00

Dimensionality reduction and manifold learning: overview and recent advances in the machine learning community

Michel Verleysen, John A. Lee, Université catholique de Louvain, Belgium

Abstract

The machine learning community has developed for quite a long time algorithmic solutions to reduce the dimensionality of data. Dimensionality reduction aims at generating a faithful low-dimensional representation of high-dimensional data, which preserves their salient and/or important characteristics, such as clusters, spatial arrangements, or topological structure. The translation of this generic principle into algorithmic solutions led to the development of a large variety of dimensionality reduction techniques having each their own strengths and weaknesses.

All dimensionality reduction methods may be used for visualization purposes if the target dimension does not exceed 2 or 3. Modern dimensionality reduction methods are technically advanced, mathematically founded, and efficient. Their properties, advantages and drawbacks have been thoroughly investigated, including the reasons why some of them perform visually better than others, and in which circumstances.

Still, most of the recent advanced techniques are not yet widely used in practice for data visualization. Despite their technical interest, they are not sufficiently known by visualization experts. They also suffer from some minor shortcomings that are unimportant issues for machine learning specialists, but that really matter for visualization experts. Tighter interaction between the two scientific communities could contribute to addressing these issues.

This course aims at providing the information visualization community with an overview of the most recent trends in dimensionality reduction developed in the machine learning community. The course will include a motivation for using nonlinear dimensionality reduction techniques, their application in visualization, a short historical perspective, a comprehensive description of the state-of-the-art techniques, and open perspectives for further development in the machine learning and information visualisation scientific communities. The course will be accessible to researchers, scientists, and practitioners with basic knowledge in mathematics.

Brief description of tutorial’s organisation, and Time allocation for major areas, the duration

The half-day course will include the following sections:

I. Introduction, basic principles, and contribution of machine learning in dimensionality reduction techniques (approx. 1/2 hour)

II. Quality assessment (approx. 1/2 hour)

III. Dimensionality reduction based on dot product and distance preservation (Multidimensional scaling and nonlinear extensions) (approx. 1/2 hour)

IV. Spectral embedding (Kernel PCA, Isomap, LLE, etc.) (approx. 1/2hour)

V. Dimensionality reduction based on topology preservation and reconstruction error (self-organizing maps and auto-associative neural networks) (approx. 1/2 hour)

VI. Dimensionality reduction based on similarity preservation (approx. 1/2 hour)

VI. Open issues for the efficient use of dimensionality reduction for information visualization (approx. 1/2 hour)

Level of tutorial

The tutorial will be accessible to researchers, scientists and practitioners with basic knowledge in mathematics. Basic knowledge of linear methods such as Principal Component Analysis is preferred but not mandatory.

Biography

John Lee and Michel Verleysen are the authors of the book “Nonlinear Dimensionality Reduction” (Springer, 2007). They both have given several tutorials and courses on the same topic at machine learning conferences (IJCNN, EANN/AIAI, JDS, ERCIM, IDEAL, COMPSTAT, etc.).

John Lee is a Research Associate with the Belgian F.N.R.S. (National Fund of Scientific Research) and Professor at the Université catholique de Louvain. He is author of about 30 publications in the field of dimensionality reduction.

Michel Verleysen is a Full Professor at the Université catholique de Louvain, and Honorary Research Director of the Belgian F.N.R.S. (National Fund for Scientific Research). He is editor-in-chief of the Neural Processing Letters journal, chairman of the annual ESANN conference, past associate editor of the IEEE Trans. on Neural Networks journal, and member of the editorial board and program committee of several journals and conferences on neural networks and learning. He was the chairman of the IEEE Computational Intelligence Society Benelux chapter (2008-2010), and member of the executive board of the European Neural Networks Society (2005-2010).

Registration-Form | IV2013 | Top | Course – Tutorial Submission