Tutorials

Tutorial Chairs: Sabine Coquillart (F), Heinrich Müller (D)

Full-Day Tutorials

T1: 3D Data Acquisition
T2: Facial Modeling and Animation
T3: Cloth Animation and Rendering
T4: Programmable Graphics Hardware for Interactive Visualization
T5: Inhabited Virtual Heritage
T6: Point Based Computer Graphics

Half-Day Tutorials

TH1: More than RGB: Spectral Trends in Color Reproduction
TH2: Modeling and Rendering of Synthetic Plants
TH3: View-Dependent Rendering for Polygonal Datasets
TH4: Geometric Data Structures for Computer Graphics

Day: Monday, Sep 2 Time: 8:30-17:30 Room: CS H2

T1: 3D Data Acquisition

R. Scopigno, ISTI-CNR, Pisa
C. Andujar, UPC Barcelona
M. Goesele, Max-Planck-Institut für Informatik, Saarbrücken
H. Lensch , Max-Planck-Institut für Informatik, Saarbrücken

3D scanners and image acquisition systems are rapidly becoming more affordable and allow to build highly accurate models of real 3D objects in a cost- and time-effective manner. This tutorial will present the potential of this technology, review the state of the art in model acquisition methods, and will discuss the 3D acquisition pipeline from physical acquisition until the final digital model.

First, different optical scanning techniques (e.g. structured light triangulation, time-of-flight approaches) will briefly be presented. Other acquisition related issues including the design of the scanning studio will be discussed and evaluated. In the area of registration, we will consider both the problems of initially aligning individual scans, and of refining this alignment with variations of the Iterative Closest Point method. For scan integration and mesh reconstruction, we will compare various methods for computing, interpolating and approximating surfaces. We will then look at various ways in which surface properties such as color and reflectance can be extracted from acquired imagery. Finally, we will examine techniques for the efficient management and rendering of very large, attribute-rich meshes, including methods for the construction of simplified triangle-based representation and sample-based rendering approaches.

Day: Monday, Sep 2 Time: 8:30-17:30 Room: MPI 019 (Visualization Room)

T2: Facial Modeling and Animation

J. Haber, Max-Planck-Institut für Informatik, Saarbrücken
D. Terzopoulos, Media Research Lab, New York University
N. Magnenat-Thalmann, MIRALab, University of Geneva
T. Vetter, Graphische Datenverarbeitung, Universität. Freiburg
V. Blanz, Graphische Datenverarbeitung, Universität. Freiburg
K. Kähler, Max-Planck-Institut für Informatik, Saarbrücken

In this tutorial we present an overview of the concepts and current techniques that have been developed to model and animate human faces. We introduce the research area of facial modeling and animation by its history and applications. As a necessary prerequisite for facial modeling, data acquisition is discussed in detail. We describe basic concepts of facial animation and present different approaches including parametric models, performance-, physics-, and image-based methods. State-of-the-art techniques such as MPEG-4 facial animation parameters, mass-spring networks for skin models, and face space representations are part of these approaches.

We furthermore discuss texturing of head models and rendering of skin and hair, addressing problems related to texture synthesis, bump mapping with graphics hardware, and dynamics of hair. Typical applications for facial modeling and animation such as speech synchronization, head morphing, and virtual aging are presented and explained.

The target audience includes, but is not limited to, students, researchers, and developers in the area of facial modeling and animation. The participants are assumed to be familiar with basic computer graphics concepts.

Day: Tuesday, Sep 3 Time: 8:30-17:30 Room: CS H3

T3: Cloth Animation and Rendering

M. Hauth, WSI/GRIS, University of Tübingen
O. Etzmuss, WSI/GRIS, University of Tübingen
B. Eberhardt, HDM Stuttgart, University of Applied Sciences
R. Klein, AG "Computer Graphik", University of Bonn
Ralf Sarlette, AG "Computer Graphik", University of Bonn
Mirko Sattler, AG "Computer Graphik", University of Bonn
J. Kautz, Max-Planck-Institut für Informatik, Saarbrücken
K. Daubert, Max-Planck-Institut für Informatik, Saarbrücken

The area of physically-based modeling is situated in the intersection of computer science, mathematics, and physics. The animation of cloth is a particularly interesting application of physically-based modeling, because it aims at fast animation solutions for rather difficult physical problems. Moreover, it addresses one of the major difficulties in creating realistic scenes with virtual actors.

The challenge of computer animation is to break down physical models for complex structures as textiles, approximate them efficiently, and run fast simulations with intelligent numerical methods. Furthermore, interactivity and collisions with other objects in the scene are challenges that have motivated much creative work over the recent years.

The range of methods proposed in literature is quite large. The techniques vary from simplified methods designed for real-time applications to sophisticated methods that were designed to reproduce measured material properties.

Rendering cloth is especially difficult because of its complex material properties. Software rendering can deal with these properties fairly easily, once they have been acquired, but remains too slow for interactive applications. Hardware accelerated rendering often provides a way to achieve interactive renderings, unfortunately complex materials aren't directly supported. We will demonstrate how interactive rendering with complex materials can nonetheless be achieved.

Day: Tuesday, Sep 3 Time: 8:30-17:30 Room: MPI 024

T4: Programmable Graphics Hardware for Interactive Visualization

T. Ertl, VIS, Universität Stuttgart
D. Weiskopf, VIS, Universität Stuttgart
M. Kraus, VIS, Universität Stuttgart
K. Engel, VIS, Universität Stuttgart
M. Weiler, VIS, Universität Stuttgart
M. Hopf, VIS, Universität Stuttgart
S. Röttger, VIS, Universität Stuttgart
C. Rezk-Salama, IMMD9, Universität Erlangen

The use of programmable graphics hardware becomes more and more important in interactive visualization applications. This tutorial will introduce the participants to programmable features of modern graphics boards and it will demonstrate how these techniques can be successfully applied to a variety of visualization algorithms. We will present volume visualization techniques for regular grids based on 2D and 3D textures and for unstructured meshes based on projected tetrahedra and we will show how pre integration significantly improves image quality. Furthermore, we will demonstrate hardware-assisted filtering and compression of volume datasets as well as visualization algorithms for flow and terrain data which also benefit from programmable graphics hardware. Finally, we conclude the tutorial giving an overview of interactive techniques for non-photorealistic rendering.

Day: Tuesday, Sep 3 Time: 8:30-17:30 Room: MPI 022

T5: Inhabited Virtual Heritage

N. Magnenat-Thalmann, University of Geneva
A. Chalmers, University of Bristol
P. Fua, EPFL, Lausanne
D. Thalmann, EPFL, Lausanne

Inhabited Virtual Cultural Heritage is a novel way of conservation, preservation and interpretation of cultural history. By simulating an ancient community within the virtual reconstructions of a habitat, the public can better grasp and understand the culture of that community.

The course will present the following concepts:

Reconstruction technology, that allows the 3D model of the ruins and the ancient forms of buildings to be constructed. This technology uses laser-scanning as well as image-based rendering techniques.
Computer Animation technology, to animate human cloth, hair and city population models, and interaction with objects.
Interaction technology, that allows the users to visualize and interact with the models intuitively and efficiently.
Augmented Reality techniques, to increase the realism of existing buildings, theaters, and sceneries etc.

Three case studies will be shown: the reconstruction of Aya Sofia and Sukullu churches in Turkey, the simulation of the Xian Terra Cotta Army, and the simulation of the Roman life in Pompei.
Necessary background: none
Potential target audience: PhD students, teachers, researchers.

Day: Tuesday, Sep 3 Time: 8:30-17:30 Room: CS H2

T6: Point Based Computer Graphics

M. Gross, ETH Zürich
H. Pfister, MERL, Cambridge, USA
M. Zwicker, ETH Zürich
M. Pauly, ETH Zürich
M. Stamminger, Bauhaus Universität Weimar
M. Alexa, GRIS, Technische Universität Darmstadt

We will present the latest research results in point based computer graphics. Affordable 3D scanning devices will be discussed and novel concepts for the mathematical representation of point sampled shapes will be presented. We will address methods for geometric processing, filtering and resampling of point models. A new framework for multiresolution modeling on irregular point clouds will be introduced. In addition, we will describe methods for high performance and high quality rendering of point models, including advanced shading, antialiasing, and transparency. The last part of the course is devoted to quickly generate point sets from complex and procedural scene descriptions.

Day: Monday, Sep 2 Time: 8:30-12:00 Room: MPI 022

TH1: More than RGB: Spectral Trends in Color Reproduction

I. Bell, University of Waterloo
G. Baranoski, University of Waterloo

Early rendering algorithms relied exclusively on three-dimensional spaces for color computation, such as RGB and CIE XYZ. Recent rendering advances use full spectral information for illuminants and surfaces, resulting in much greater accuracy and realism. These expensive computations can be wasted, however, if ad hoc methods are used to adjust the final image on the monitor, in film, or in print. Inefficiency and inaccuracy can be avoided with some knowledge of device gamuts and color reproduction algorithms.

This course follows spectral data through the graphics pipeline, examining issues of rendering, color science, perception, gamut mapping, and color management. We conclude with a discussion of trends and open problems in managing spectral data for accurate color reproduction. Participants will learn not only the theoretical background of color and spectral reproduction, but practical guidelines often omitted in technical papers.

Day: Monday, Sep 2 Time: 8:30-12:00 Room: MPI 024

TH2: Modeling and Rendering of Synthetic Plants

Oliver Deussen, Dresden University of Technology

Many computer generated images contain plants or show outdoor scenes. Therefore it is a challenging task to model and to render plants and whole ecosystems efficiently. Over the past years a variety of plant modeling algorithms have been proposed that use different generation schemes and provide different model quality.

In the tutorial the algorithms are classified into rule-based and procedural approaches, a number of them is presented. Methods of combining plants to ecosystems are given. Rendering of ecosystems will be described using polygonal techniques and also image-based rendering methods. In the last part the generation of non-realistic images of plants will be discussed with applications in architecture and landscaping.

Day: Monday, Sep 2 Time: 14:00-17:30 Room: MPI 022

TH3: View-Dependent Rendering for Polygonal Datasets

J. El-Sana, Ben Gurion University, Israel
L. De Floriani, University of Genova
E. Puppo, University of Genova
A. Shamir, The interdisciplinary center, Herzelia

Multi-resolution hierarchies and view-dependent rendering have been introduced to enable various levels of detail to seamlessly co-exist over different regions of the same surface. These levels of detail depend on parameters such as view location, illumination, and speed of motion and are determined per-frame.

This tutorial will focus on describing techniques to construct multi-resolution hierarchies for geometric objects, and utilizing these hierarchies to accelerate rendering and transmission of large polygonal datasets. It is intended for those who have an understanding of the basics of 3D graphics and analysis of algorithms. The goal of the tutorial course is to expose both students and professionals to various advanced techniques and algorithms to accelerate the rendering of large polygonal datasets through the use of multi resolution hierarchies.

The tutorial shall start by presenting level of details and multi-resolution hierarchies in the general framework, then it covers the recently developed view-dependent rendering approaches which include multi-triangulation, generalized view-dependent rendering, and multi-resolution hierarchies for dynamic objects and scenes. Lastly, it will introduce advance techniques such as view-dependent rendering over network and integrating view-dependent rendering with occlusion culling.

Day: Monday, Sep 2 Time: 14:00-17:30 Room: MPI 024

TH4: Geometric Data Structures for Computer Graphics

G. Zachmann, AG "Computer Graphik", University of Bonn
E. Langetepe, AG "Computer Graphik", University of Bonn

In recent years, methods from computational geometry have been widely adopted by the computer graphics community.

This tutorial aims at familiarizing practitioners in computer graphics with a wide range of data structures from computational geometry. This will enable the attendants to select the most suitable data structure when developing computer graphics algorithms. In particular, they will be able to readily recognize a problem when it can be mapped onto one of the methods presented in this tutorial.

Each geometric data structure will be presented in the following manner: first, the data structure will be described and some of its fundamental properties will be highlighted; after that, efficient computational geometry algorithms based on the data structure will be presented; and finally, a number of recent representative and practically relevant algorithms from computer graphics will be described, highlighting the utilization of the data structure in a creative and enlightening way.

The tutorial will focus on algorithms and data structures which have proven to be versatile, efficient, fundamental, and which can be readily implemented by experienced practitioners in the field. This will ensure that the audience will acquire knowledge they can utilize immediately for their work.

The intended audience are practitioners working in 3D computer graphics (VR, CAD/CAM, entertainment, animation, etc.). They should be familiar with the basic principles and the type of problems in the area.

The audience comprises also students from both computer graphics and computational geometry, possibly working on a master or PhD thesis, who want to get a feeling for the potential and possibilities obtained by applying computational geometry to computer graphics problems.

This page is part of the Eurographics 2002 web site.
Please contact the Webmaster if you have any questions or comments concerning this page.
Page last edited on Monday, 26-Aug-2002 14:30:46 CEST.

Eurographics 2002 is sponsored by

Imprint-Dataprotection