2 /2007 Vol. 19

COMPUTER GRAPHICS

Reports on Computer Graphics

Seminare, Kongresse, Workshops im

Zentrum für Graphische Datenverarbeitung www.zgdv.de Windows 2003 Server Grundlagen 03.-05.09.2007 Installation, DNS, Active Directory (AD), Gruppenverwaltung, Sicherheit, Backup/Restore

Adobe Photoshop Grundlagen 01./02.10.2007 Paletten, Adobe Bridge, Bilder bearbeiten und optimieren, Arbeiten mit Pfaden und Ebenen, Text und Farben

Softwaretest in der Praxis 04./05.09.2007 Testplanung, Risiken, System- und Abnahmetest, Reviews, typische Probleme

Repräsentation von Geodaten mit der Geography Markup Language GML3 09.10.2007 XML , Namespaces, Xlink, XML Schema, Zuschnitt von GML3 auf Applikationen, ISO 19107 und Umsetzung, AFIS/ALKIS®/ ATKIS-Modell der AdV, CityBML

Adobe InDesign - Grundlagen 10./11.09.2007 Benutzeroberfläche, Textbearbeitung, For® mate, Zeichenwerkzeuge, Tabellen, Ebenen ALKIS Einführung und Modellierung 10.10.2007 Workshop 3D Stadtmodelle: AnwenALKIS®-Struktur der AdV, Migration, dungen und Forschung Grundlagen der NAS, GeoInfoDok 12.09.2007

NAS Die neue Schnittstelle im LiegenSoftwareübersicht, Geodatenbanken, Datenqualität, Visualisierung, WebServices, schaftskataster 17.10.2007 Datenaustausch ATKIS®-ALKIS®-AFIS®-Referenzmodell, Web 2.0 Intensivseminar Abbildung von Geschäftsprozessen, Ablei14.09.2007 tung, Referenzdatenbestand, Analysen Blogs, Podcast, Social Software, Folksonomy, Web als Plattform, WEB 2.0, Ausblick Darmstädter Kongresse WEB 3.0 Java 5 Tiger Grundlagen 17.-19.09.2007 Konzepte, Eclipse, Programmierung, Streams, Exceptionbehandlung, Java, Applets, Assertions, Team-Entwicklung

Semantic Web und Wissenstechnologien

SVG - Scalable Vector Graphics 17./18.09.2007 XML, SVG-Viewer, PlugIns und nativ, Text, Transformationen und Pfade, ElementeGruppen, Masken und Filtereffekte, Scripting, Animationen, Webmapping Moderne Softwareentwicklung 24.-26.09.2007 Sprachen; Prinzipien: Single-Responsibility, Package Coupling u.v.m.; Pattern: Abstrakte Fabrik, Besucher, Kompositum...; Prozesse: Extreme Programming, MDA Adobe Acrobat PDF 24.09.2007 Acrobat Distiller, Dateien bearbeiten, Formularfelder, Adobe Life Cycle Designer, PDF und HTML, Adobe Catalog, Preflight LaTeX Professionelle Textverarbeitung 01./02.10.2007 Übersetzung, Gleitobjekte, Formelsatz, Einbinden von Graphiken, Index und Glossar Adobe InDesign 04./05.10.2007 Navigation, Musterseiten, Seitenlayout, Nummerieren, Verketten, Tabellenfunktion, Bibliothek, Farbauszüge, Im-/Export

18. Oktober 2007 in Darmstadt

Adobe Illustrator 08.-10.10.2007 Im Rahmen der Darmstädter Kongresse Objekte erstellen und bearbeiten, Pfade, in Planung: Flächen/Konturen, Verläufe, Stilpalette Barrierefreie PDF-Dateien erstellen und ® • 6. Kongress XML und ALKIS optimieren Videoschnitt mit Adobe Premiere Pro • Mobiles GIS 17.09.2007 15.10.2007 • 3. Kongress 3D Stadtmodelle BITV und PDF, Grenzen, PDF oder HTML?, Grundkonzepte, Digitalisierung, Schnitt, Workflow, Tagged PDF, Erstellung aus MS Farbkorrektur, Betitelung, Videoausgabe Office, bestehende PDF`s optimieren Cisco PIX Firewall - Einführung 24.-26.09.2007 Cisco PIX Security Appliances Produktfamilien, Firewalls Designs, PixOS, SSH, Telnet, HTTP, AAA, sichere DMZ, Q&A

konziel, agentur für kommunikation, B. Lukacin; +49 / 6151 / 155 146

COMPUTER GRAPHIC 2/2007, Vol. 19

topics

Anmeldung ZGDV Darmstadt Aus-, Weiter- und Fortbildung Hugo Kopanitsak, Alexandra Ohly Phone: ++49 (0) 6151/155-160+163 E-Mail: [email protected]

Anmeldung ZGDV Rostock Aus-, Weiter- und Fortbildung Eva Mahnke, Ute Woitzel Phone: ++49 (0) 381/4024-159 E-Mail: [email protected]

http://www.zgdv.de/

http://www.zgdv.de/

Technische Universität Darmstadt, Fachgebiet Graphisch-Interaktive Systeme (TUD-GRIS) Technische Universität Darmstadt, Interactive Graphics Systems Group Zentrum für Graphische Datenverarbeitung e. V. (ZGDV) Computer Graphics Center Fraunhofer-Institut für Graphische Datenverarbeitung (IGD) Fraunhofer Institute for Computer Graphics

Reports of the INI-GraphicsNet Editorial Office: COMPUTER GRAPHIC topics Fraunhoferstrasse 5 64283 Darmstadt Germany Phone: +49 (0) 6151/155-146 Fax: +49 (0) 6151/155-446 E-mail: [email protected]

Publisher: Prof. Dr.-Ing. Dr. h.c. mult. Dr. E.h. Hon. Prof. mult. José L. Encarnação

Art Direction: Detlef Wehner, Ralph Klepper, Bernad Lukacin

Issue Editors: Dr. Ing. André Stork Dr. Stefan Göbel Editor: Bernad Lukacin

Translation: Elfriede Fitschen »COMPUTER GRAPHIC topics« is published six times a year. All rights reserved. Not to be reprinted without approval of the editor.

INI-GraphicsNet Join the new technology Computer Graphics is one of the key technologies of a modern information and knowledge society. The INI-GraphicsNet develops market-oriented, state of the art technology to foster and to support the innovation process of enterprises as well as the social development. Numerous businesses use our know-how to implement sustainable products and services. We achieve this with, for example: – – – – – – – – – – – – – – –

3D Interaction and Visualization Agent Technologies Animation Augmented Reality Computer Supported Cooperative Work (CSCW) Database Services Geometric Modeling / CAD-Model Graphical Information Systems (GIS), Facility Management GUI / Interaction Technology Image Analysis, Image Quality Imaging Internet- & Intranet-Solutions IT-based Learning and Training Medical Data / Image Processing Mobile Computing Technology

– – – – – – – – – – – – –

Modeling and 3D-Reconstruction Multimedia and Hypermedia Multimedia User Interfaces Perceptual Computing Product Data Management Rendering Scientific Visualization Secure Image Communication Semantic Modeling System Integration Telecommunications Ubiquitous Computing Usability and Utility Engineering Technologies & Methods – Virtual Prototyping – Virtual Reality

Our expertise allows us to work on a multitude of industry-related topics which include, amongst others: – – – – –

eApplications eServices eBusiness Medical Information Technology IT Security and IT for the security in our society – Visualization and Interaction in traffic technology and traffic telematics

Computer Graphics Computer graphics is the technology with which pictures, in the broadest sense of the word (synthetic graphics as well as grayscale and color images), are captured or generated, presented, manipulated, digitally processed in the appropriate form for the respective application and merged with other, nongraphical application data. Computer graphics also includes the computer-supported integration and manipulation of these pictures with

– – – –

Ambient Intelligence Games and Edutainment Usability and Utility Engineering Software for the product and production development

Contents CAD Systems Become Steadily More Important for Industrial Applications

5

Innovative Visualization Techniques Facilitate the Work with Complex Data 7 Virtual Reality Facilitates and Accelerates Product Development

10

Augmented Reality – A Mix of Real and Virtual World Supporting Maintenance and Training

13

Optimized Simulations Accelerate the Product Development Process

16

The New Main Research Areas of Fraunhofer IGD

18

20 Years of Computer Graphics – an Interview With Prof. Dr. José L. Encarnação and Prof. Dr. Dieter W. Fellner

20

Go Digital: Toss Pin Boards and Flip Charts

22

INSCAPE: Interactive Storytelling for Creative People

25

GameDays 2007

27

RUBRICS other kinds of data, such as audio, speech and video (to create multimedia systems) as well as corresponding advanced dialog and interactive technologies. Concepts which characterize the important topics of computer graphics are, to name a few, visualizing information, visual data mining, visual computing, Virtual Reality (VR), Augmented Reality (AR), interactive Internet services and secure image transmission and communication.

News Events StudINI Graduations Study and Diploma Theses

29 36 45 46 48

EDITORIAL

Dieter W. Fellner

When, in 1987, the Fraunhofer Institute for Computer Graphics Research IGD was founded the World Wide Web was still a vision of some scientists for the future, Virtual and Augmented Reality was better known from science fiction films than from industry, the buildings and machines were designed by drawings only interpretable by experts, and product developments had to be tested again and again with real prototypes. That all this has changed, is also owed to the scientists of Fraunhofer IGD. For more than twenty years they have been working on transforming figures and formulas into images, thus enabling many applications we now take for granted. So, for example, Virtual Reality could only be established in the industry, because the scientists found technological solutions to create complex threedimensional worlds in real time. Also medical applications like the computer tomography and three- or fourdimensional ultrasound pictures were unthinkable without the developments of computer graphics. From the very beginning, one of the core markets for developments of computer graphics was the industrial sector. Those companies were the first to use comprehensively and area-wide CAD systems in the construction process, to use the first Virtual Reality systems, and it is the industry that will most benefit from the current developments around Simulated Reality. Especially very competitive industries like the automobile sector but also the machine building industry, so successful in Germany, experience increasingly shorter product life cycles. New technologies promising shorter product development times at lower costs and a higher quality will therefore play an increasing role.

4

CG

topics 2/2007

How development, production, maintenance, and servicing processes have changed within the last two decades thanks to computer graphics developments shall be described in our special supplement pages of this topics edition on the occasion of the 20th anniversary. In five contributions we will give you an overview on our work of the last twenty years. We will show you how all this began: with developments mainly dealing with the CAD applications new at that time. So Fraunhofer IGD has considerably contributed to define CAD standards, which, in the first place, allowed for the triumphant success. Already in the first few years, the scientists also developed new visualization technologies and interaction tools more in line with the user habits than the programs. An important step to increase the acceptance of such systems. These developments have persisted till today. A new dimension of visualization was then offered by the Virtual Reality which, from the beginning of the 1990s, has started its triumphal procession from the research laboratories to the industry, slowly first, then quicker and quicker. Mainly in the automobile industry Virtual Reality is nearly standardly used today, for instance to carry out design reviews. The more the developments in the field of Virtual Reality had advanced, the more obvious it became that also for documentation and instruction purposes new solutions must be included. So, in the late 1990s, first tests were executed with the socalled Augmented Reality. Also this time, the automotive industry was the driving force of the developments. In addition to the visualization of complex models and the interaction with those objects the overlay of virtual models with real objects has been in the focus of Fraunhofer

IGD’s scientists. Another step into this direction has been done by the scientists with their work dealing with Simulated Reality. Here applications have come into being since 2005 supporting the engineers in the complex search for the optimal product design. So far the overview of our past and current work. But you will also find a little foretaste of our future topics. So our research work will increasingly deal with semantics in the model building domain, with library questions in the context of multi-dimensional objects, as well as with the confluence of computer graphics and computer vision. In addition to our special subject, the 2nd Computer Graphics topics 2007 is as usually also giving an insight into some current projects of the INI-GraphicsNet members. Furthermore, we are reporting on news and events of the last months and presenting our scientists who have received a PhD in the last months. We wish you an interesting reading.

CAD Systems Become Steadily More Important for Industrial Applications Julia Mayer

German Abstract Bereits zu Beginn der Arbeiten des Fraunhofer IGD bildeten Forschungsarbeiten zur Vereinfachung, qualitativen Verbesserung und Beschleunigung industrieller Entwicklungs- und Produktionsprozesse einen der Schwerpunkte der wissenschaftlichen Arbeiten des Instituts. Im Gründungsjahr 1987 waren Industrieunternehmen und vor allem Design- und Architekturbüros noch weit davon entfernt, standardmäßig CAD-Systeme in ihren Konstruktions- und Entwicklungsprozessen einzusetzen. Einem weitverbreiteten Einsatz der Systeme standen vor allem die mangelnde Standardisierung, fehlende Schnittstellen zwischen den verschiedenen Systemen sowie die kaum vorhandene Anpassung der CAD-Systeme an die Anforderungen der Benutzer im Weg. Seit seiner Gründung arbeitete das Fraunhofer IGD daran, diese Schwierigkeiten zu beseitigen und den Weg für einen flächendeckenden Einsatz von CAD-Systemen in der Industrie zu ebnen.

From the early days of Fraunhofer Institute for Computer Graphics Research IGD research on the simplification, qualitative improvement, and acceleration of industrial development and production processes was one of the main areas of the institute’s scientific work. In 1987, the year of foundation, industrial enterprises and, above all, design and architecture firms were far from a standard use of CAD systems in their construction and development processes. A widespread use of the systems was barred by a lacking standardization, by missing interfaces between the different systems, and by the fact that CAD systems were scarcely adjustable to the needs of the users. Since its foundation, Fraunhofer IGD has worked on resolving these difficulties to clear the way for an area-wide use of CAD systems in the industry. Among these activities were for example the following projects executed in the first years of the institute: Generation of a GKS-compatible Graphics System with PHIGS Functionality Independently from the research activities in the field of CAD programs, standards for computer graphics had to be established, in order to pave the way for a wide usage of those technologies. In the year 1987, the two 3D graphics systems GKS-3D and PHIGS were near standardization. The portability of graphics applications between both systems, however, was obstructed by a missing upward compatibility of GKS3D to PHIGS. In April 1987, Fraunhofer IGD started to solve this problem and to push ahead the implementation of PHI-GKS. The aim of the implementation was to provide an environment allowing to run GKS and

GKS-3D, as well as PHIGS programs, to provide more functionality in one system. Integration of the Constructive Engineering Geometry into Geometric Modeling In the early days of Fraunhofer IGD the user interfaces of many CAD systems were oriented by their own problems rather than the users’ workflows and requirements. Computer and user world diverged widely. The systems forced their users, having poor or no informatics knowledge at all, to change their work methods. So it was nearly impossible to achieve a high user acceptance, which was one of the major problems impeding the general use of CAD systems. To solve this problem, to allow the users a more intuitive work with the CAD systems and so increase the acceptance of the systems researchers of Fraunhofer IGD developed a »frontend« modeling system. The constructive geometry methods were offered within the dialog module and formed the communication interface to the user. To create a user interface as clear, solid, and simple as possible the developers attached great importance to the following: – Functions grouped according to their semantics – Graphics-oriented dialog using symbols – Auxiliary texts reporting the finished or expected actions – Tests before any action of the user as a protection against errors – Reduction of the work with keyboard – Sufficient information on the screen about the system status – Use of a high-resolution screen

1987 1987 CG

topics 2/2007

5

Another step on this way to a better usability of CAD systems was done by Fraunhofer IGD from the year 1991 on, when they defined industrial design as a CAD application area. In this connection the scientists dealt with generating tools that allow users to draft ideas with pen and paper as usual. It was the aim to allow the sketchy input of any geometry elements by pen and tablet also with 3D CAD systems. With the CyberStilo the institute succeeded in doing so some years later. CAD Reference Model At the end of the eighties the situation of CAD systems was marked by a multitude of application technologies, a rapid spread, and a considerable innovation speed. This resulted in incompatible systems, inconsistent and incomparable structures, and a different and ambiguous terminology. So there was a need for a reference model covering across the application areas all aspects of the CAD technology and serving as a system scheme and basis. A work group formed within the department 4 of the Gesellschaft für Informatik GI e.V. (society for computer science) should give a first draft of the CAD reference model. Fraunhofer IGD too was active in this project. The institute’s task was to further develop the intention in projects. For this aim the scientists have specified the architecture for the reference model and prepared a study about possible formal specification languages. The joint project CAD Reference Model was funded by the former Federal Ministry of Research and Technology (BMFT) after a start-up phase from October 1991. Presentation in STEP Respectively Graphic Information in the CAD Exchange Format One of the institutes key activities in the first few years was the contribution to the development of the international standard STEP. STEP stands for »standardization of the external representation of product definition data«. The standard which is still used today was a great leap forward for the use of CAD data. The generated interface allowed

Figure 1: Processing of CAD vehicle parts.

Figure 2: User interface of ARCADE - a collaborative CAD tool.

an exchange of product defining data, so-called CIM data, between CAD systems as well as between CAD systems and upstream or downstream CA islands like CAP, CAM, or FEM. The objective was to exchange via STEP not only technological but also organizational product data for the first time. So STEP was a conceptual extension of the common national standards VDA-FS, IGES, and SET, all generally confined to the exchange of the geometry data of the product. Furthermore, with STEP the prerequisite for digital product lifecycle management had been created. Fraunhofer IGD’s task in the scope of the STEP project concentrated on two areas: presentation and geometry. In the field of presentation it was about defining a neutral interface containing the instruction for generating a twodimensional attributed view of the 3D product model. For this aim a presentation sequence for the geometry and labeling was defined. A mechanism

was established allowing to assign presentation attributes to the geometry and the labels on each level of the hierarchy and, at the same time, controlling the inheritance of the presentation attributes. In the field of geometry, the cooperation focused, on the one hand, on the correction of faulty entity definitions and, on the other hand, on restructuring the existing entities to facilitate the formation of STEP power stages. The internal research activities of Fraunhofer IGD of the years 1987/88 in STEP were funded by the BMFT in the scope of the project »KCIM in DIN« from March 1988. So, in the first instance, the urgently required actions were formulated and the activities in the areas considered important reinforced in cooperation with other research institutions. Till now, STEP has been one of the most important standards for the description of product data.

1991 1988 6

CG

topics 2/2007

Innovative Visualization Techniques Facilitate the Work with Complex Data Julia Mayer

German Abstract Die Visualisierung war und ist eines der Kerngebiete des Fraunhofer-Instituts für Graphische Datenverarbeitung schlechthin. Unter Visualisierung ist dabei die Umsetzung umfangreicher Datenmengen in Bilder zu verstehen, anhand derer die in den Daten verborgenen Informationen erkannt und extrahiert werden. So können die Informationen schneller, verständlicher und im Zusammenhang mit weiterführenden Erkenntnissen vermittelt werden. Je größer die Bedeutung von CAD und digitalen Berechnungen für die industrielle Produktentwicklung und -konstruktion wurde, desto mehr nahm auch die Bedeutung der anschaulichen Visualisierung dieser Prozesse zu. Und so rückte die Visualisierung komplexer Prozesse mehr und mehr in den Fokus der Wissenschaftler. Doch nicht nur die »Verbildlichung« komplexer Daten spielte und spielt dabei eine wichtige Rolle, wenn sich die Fraunhofer IGD Wissenschaftler mit Visualisierungstechnologien befassen. Auch die Entwicklung neuer, intuitiverer Eingabewerkzeuge, die sich mehr an der Arbeits- und Lebenswelt der Nutzer orientieren, gehört seit Anfang der 90er Jahre zu den Forschungsthemen des Instituts.

From the foundation of the institute till today, visualization has been one of the core areas per se of the Fraunhofer Institute for Computer Graphics Research IGD. Visualization is understood as the transformation of voluminous amounts of data into images that serve to recognize and extract the information hidden in the data. So the information can be transferred more quickly, more intelligibly, and in relation to further findings. The more CAD and digital calculations for the industrial product development and construction grew in importance the more a clear visualization of these processes and products also grew in importance. Consequently the visualization of complex processes has more and more moved into the focus of the scientists. But not only the »visualization« of complex data has played an important role when scientists of

Fraunhofer IGD have dealt with visualization technologies. Also the development of new, intuitive input tools, that are more geared to the working world and the social environment of the users, have been among the research topics of the institute since the early nineties. Interactive Visualization of Volume Data (InViVo) Goal of the project started in 1992 was to raise the speed and flexibility of the visualization of big, threedimensional, scalar fields. Such data fields can be found in many scientific, technical, and medical applications like with 3D measurements of temperature, pressure, density, or concentration or with imaging diagnostic techniques like CT, MRI, and 3D ultrasound. InViVo was the first project to integrate many new developments of previous years concerning lighting models for volume data

Figure 1: CASUS

1992

1992 CG

topics 2/2007

7

and the faultless scanning of discrete data fields. In the course of the following years InViVo was systematically developed to support medical applications. It was a breakthrough when, in 1994, Prof. Dr. Georgios Sakas for the first time successfully achieved a three-dimensional visualization of an embryo in the womb on the basis of ultrasound data. In the year 1997 the software license was sold to MedCom, a spin-off of Fraunhofer IGD founded by Prof. Sakas. Since then, they have successfully realized medical applications on the basis of InViVo. VIS-A-VIS: Development of a Visualization Toolkit for Parallel Systems Especially in the technical/scientific area like fluid mechanics, medical research, or material research immense amounts of data accrue. To interpret these data without a graphical visualization was extremely difficult and time-consuming. To find a solution for this problem was the goal the researchers of Fraunhofer IGD pursued with the development of the VIS-A-VIS toolkit. This visualization system was based on highquality computer graphics methods for 3D presentation which – to allow an efficient interactive visualization and to counter the growing real-time requirements in visualization – should be as far as possible parallelized for implementation. One of the biggest problems the scientists had to face was the high computing intensity of the direct graphical presentations of measurement and simulation data defined in the three-dimensional space in high quality. To solve this problem an improved front-to-back (FTB) projection algorithm was developed serving to compute the presentations. Their resolution could be selected independent of the resolution of the dataset. This method met the accuracy requirements, but was optimized for the opaque visualization of volume data. Then, in 1992, this module for the visualization of volume data was parallelized on a multi-processor workstation.

Figure 2: User interface of the ConeptViewer.

In the same year, also the extensive but efficient ray casting method for the visualization of volume data was ported onto a multi-processor system (MANNA) developed by the Gesellschaft für Mathematik und Datenverarbeitung (society for mathematics and data processing) GMD. With this implementation the performance of the massive parallel computer system and the selected distribution strategy could be demonstrated. In the scope of the VIS-A-VIS toolkit also new interaction devices like a data glove or a trackball were developed enabling to operate several dimensions at the same time and offering a more intuitive handling of the systems. In addition to the visual implementation of numeric files the scientists also worked on the transformation of numeric data into audible signals. By combining visualization and sonification they wanted to increase the informative value of the presented information and to offer additional information.

1994 8

CG

topics 2/2007

ISVAS – Interactive Software for a Visual Analysis A possible application for the VIS-A-VIS toolkit was shown by the ISVAS project. ISVAS, a visualization system for the analysis of the results of finite element calculations developed at Fraunhofer IGD, was used in 1992 for example to depict numerically calculated flow fields in a useful way. This technology provided considerable improvements especially for complex geometries of the flow channel and very large datasets, required for example for the simulation of injection processes in car cylinders. Already in 1992, long before the Internet became a mass medium, the Fraunhofer IGD scientists adopted in ISVAS also an interface for an online visualization. MoVi – Mobile Visualization In 1995 scientists of Fraunhofer IGD began to reflect about how the mobile end devices could be used to combine mobility and information supply. Goal of the joint project MoVi, Mobile Visualization, funded

1995

1995

a unique depth of focus and 3D rendering. Therefore, the system that is meanwhile marketed as a product is ideal for the presentation of complex products and processes.

Figure 3: The HEyeWall®: a worldwird unique presentation system.

by Deutsche Forschungsgemeinschaft (German Research Foundation) DFG, was to develop mechanisms for the visualization of multimedia information on mobile, networked computers in a user-friendly, comfortable form. What today mainly applies to mobile phones was at that time one of the greatest challenges for developments for the first laptops: their limited local resources and their limited performance. Also the »comparably tiny display« (annual report 1995) was one of the special challenges the scientists had to face. Internet-based Visualization In the mid-1990s the Internet was still a medium for some technophile specialists. But at Fraunhofer IGD some first research work started on visualization technologies of the WWW. So technologies of the WWW-based, distributed visualization of scientific/technical data should be developed and evaluated in a pilot project by order of and in cooperation with two partners. The goal was to realize a wide range of visualization solutions with Internet technologies to be able to quickly, flexibly, and individually provide experts and laypersons with graphical representations of data. For this aim

different visualization approaches for the use in the Internet and the Intranet were prepared and implemented prototypically. Using technologies new at that time like Java, CGI connections, and VRML individual visualizations could be realized on a fast special computer of the visualization provider and then displayed on a WWW client of the user. From late 1996 the visualization solutions developed in the project were tested internally. The HEyeWall® – A Presentation System Unique Worldwide Until a few years ago it was not possible to quickly and easily represent complex products and processes in real time. There were no large-size presentation systems to display satellite pictures, vehicles, simulations of air traffic, or complex information in an extremely high resolution and, thus, in every detail. This gap was closed by the presentation system HEyeWall® presented for the first time to the public in 2004. Its resolution of 6144 x 3072 pixels lies above that of the human eye. So far the viewer had to accept blurred contours, pixels, and muddy colors, but with the HEyeWall® he can now examine from any position details in

1997 1996 CG

topics 2/2007

9

Virtual Reality Facilitates and Accelerates Product Development Julia Mayer

German Abstract Mitte der 90er Jahre war der Einsatz von CAD- und CAE-Systemen in der Automobilindustrie weit genug vorangeschritten, um eine Basis für die Umsetzung erster virtueller Prototypen zu bilden. Ziel des Einsatzes von VR-Technologien in der Industrie war es, den Bau physikalischer Prototypen zu reduzieren oder ganz zu vermeiden und durch virtuelle Modelle zu ersetzen, um so die Produktentwicklung zu beschleunigen und Kosten zu reduzieren. Erste Arbeiten im Bereich der Virtuellen Realität begannen am Fraunhofer IGD bereits 1990. Im Laufe der Jahre entwickelte das Institut mit einem VR-Toolkit Systembausteine für alle Bereiche der Virtuellen Realität, betrieb ein VR-Demonstrationszentrum und installierte die weltweit erste 5-Seiten CAVE. Durch seine Forschungs- und Entwicklungsarbeiten auf diesem Feld trug das Fraunhofer IGD maßgeblich dazu bei, die Virtuelle Realität in die Industrie zu tragen.

In the mid-1990s, the use of CAD and CAE systems in the automotive industry was advanced enough to provide the basis for an implementation of the first virtual prototypes. Using Virtual Reality (VR) technologies in the industry the construction of physical prototypes should be reduced or even avoided, thus speeding up the product development process and reducing the costs. Already in 1990, Fraunhofer IGD started to do research work in the field of Virtual Reality. Over the years, the institute has with a VR toolkit developed system modules for all areas of Virtual Reality, run a VR demonstration center, and installed the worldwide first 5-sided CAVE. The research and development work performed at Fraunhofer IGD has considerably contributed to carry Virtual Reality into the industry. Here some examples of Fraunhofer IGD’s activities in the field of Virtual Reality: Demonstration center »Virtual Reality« In the framework of an SME program of Fraunhofer-Gesellschaft the Fraunhofer demonstration center »Virtual Reality« was set up in1993 in cooperation with the Fraunhofer Institutes IAO, IBP, and IPA. The goal was to realize over a periode of five years a demonstration and competence center for VR within the Fraunhofer-Gesellschaft. For this purpose demonstration centers were built up in the four institutes. In the focus of the demonstration center at Fraunhofer IGD were especially small and medium-sized enterprises which should benefit from this technology. Fraunhofer IGD was the

1990 10

CG

topics 2/2007

only German institute at that time offering solutions in the whole system area and giving support in the design of applications, as consultants and usage planners. Before the demonstration center’s foundation Fraunhofer IGD had already generated system modules for all areas of Virtual Reality, the so-called VR Toolkit. Also the VIS-AVIS rendering system of Fraunhofer IGD conceived as a universal system could be used. This system could handle different renderers from a common object handler. Furthermore, additional modules could be developed based on the real-time mode implemented in VIS-À-VIS, which allowed the realization of applications from the virtual world. One of the projects realized in the demonstration center was the »Virtual Design«. Here a special VR environment for the import and presentation of CAD model data was generated. It was based on the VR Toolkit of Fraunhofer IGD and supported the preparation and conversion of a multitude of CAD data types so that all existing models could be easily integrated. As the development of the needed VR infrastructure means high investment cost for enterprises, the demonstration center could at that time already offer the execution of presentations as a service. So CAD planning data could virtually manifest and be experienced by anybody. From 1996 the lab was extended by a 3-sided CAVE consisting of several stereographic projection sides (floor and side walls). For the first time the user was present with his own body in the computer-generated world.

1993

Figure 1: Virtual Design in the 5-sided CAVE.

VR System for the Visualization of Vehicle Structure Components In 1996, Fraunhofer IGD, by order of BMW AG, designed one of the first VR systems that were integrated into the production process. It was the aim to develop a VR system for the visual analysis of vehicle structure components. Based on a demonstrator developed in the previous year the data export from the »Virtual Workshop« and the import into the »Virtual Design« system was coordinated with Tecoplan und realized within the project. Using a graphical surface the user was able to import the geometries of the structure components, process them, start the VR system to view the selected components with the input and output device BOOM (Binocular Omni Oriented Monitor), and perform different actions.

1996

Virtual Reality for Assembly and Mounting Simulation Background of the project, also realized by order of BMW AG, was that at that time the mountability and serviceability of components in the automotive industry could only be tested by a very cost- and timeintensive model-building. The prototypes served design engineers to test for example the assembly processes of the planned components, to discuss arising problems, and to modify constructions which had then again to be built as prototypes. To considerably optimize these inefficient planning steps Fraunhofer IGD developed a VR system for the simulation of assembly and mounting. This included, on the one hand, the connection of the VR systems to CAx and PDM environments with BMW and, on the other hand, the use of special features and interaction components in the VR system

itself. Aim of the project was to directly import the constructed data from the CAD environment into a VR system, ensuring that all those questions were mapped in the virtual environment that previously could only be answered by means of real prototypes. In the final BMW benchmark tests the system of Fraunhofer IGD showed clear advantages compared to other VR systems. The first 5-sided CAVE Worldwide With the opening of the new Fraunhofer IGD building in 1997 also the worldwide first 5-sided CAVE was installed. Due to the projections on 5 sides the disturbing shadows of 3- and 4-sided CAVEs on the floor projection could be eliminated. Two-handed interaction possibilities, the user’s full freedom of movement, the 1:1 representation of the objects, and the possibility to discuss the results with other engineers, make the CAVE the optimal output device for digital prototyping. Therefore, most of the CAVE projects were and are realized in the scope of digital prototyping with enterprises from the automotive and shipbuilding industry and from plant and aircraft construction. IVIP – Virtual Product Development The idea that a general digitization of the product development process is one of the keys for increasing the competitiveness of enterprises was the starting point of the lead project IVIP funded by the German Federal Ministry of Education and Research. From 1998, 50 partners worked within the project on questions arising from the use of digital prototypes in the product development. The goal was to provide an integrated overall solution allowing a completely digitized product development. Fraunhofer IGD was involved in two sub-projects: The software connection of a force feedback system for virtual environments to the VR system »Virtual Design II« and the development of digital testbeds,

1997 CG

topics 2/2007 11

supporting hybrid testing of digital and physical prototypes. The aim of the latter was to create an efficient arrangement of tests covering the phases planning, installation, realization and analysis of tests.

Figure 2 (right) and 3 (bottom): Two examples, rendered with OpenSG in realtime.

OpenSG – The Open Source Scene Graph The development of OpenSG began in 1999 with the aim to elaborate a freely available scene graph featuring – in addition to a high rendering performance – the following: – Portability – Multi-threading support – Support of several graphics pipelines – Support of clusters – Extensibility – Flexibility The kernel of OpenSG was developed at Fraunhofer IGD. Today, OpenSG is the basis for numerous VR projects and is continuously advanced by a worldwide developer group. CAx/VR Integration Tools for the Support of DMU and VP Processes Around the turn of the millennium the development of digital mock-up (DMU) and virtual prototyping (VP) had technically advanced so far that they were in fact ripe for the practical use in the industry. This process was, however, prevented in many areas of the producing industry by the very time-consuming transition from construction, simulation (CAD/CAE), and product data management (PDM) into Virtual Reality, by a poor computer support, and lacking internal know-how. One of the core activities of Fraunhofer IGD since the turn of the millennium has therefore been the realization of tools for a better integration of VR systems into the production process. Goal of the different projects in this domain is an effective support of the processes to digital mock-up and virtual prototyping. Among others, the joint project »Digital Mock-Up Visualization in product conception and down-

stream processes« (DMU-VI) has contributed to this aim by introducing DMU tools for improving a consistent, supported digital product development and an optimization of the product development processes. The task of Fraunhofer IGD within this project was the conception and specification of the DMU-VI system architecture as well as the development of the CAD server being the link between an engineering environment (CAD/CAE/PDM) and VR for the immersive real-time visualization of complex product models. Also the analysis of relevant standards as to their role in and their impact on DMU-VI was part of the institute’s activities.

2000 12

CG

topics 2/2007

In the project »Integrated VR Interface« the CAD server prototypically realized in cooperation with vrcom GmbH in the DMU-VI project was brought to product maturity until March 2000. The heart of the VDDP (Virtual Design Data Preparation) is a graphical editor for the preparation of modules and components allowing to import 3D CAD models in a parametric representation, to tessellate them in a suitable way, and to optimize them in their representation quality. Furthermore, model errors like miscalculated surfaces or multiple surfaces can be corrected. Another elementary component of VDDP is a converter allowing to import CAD models from CATIA (a CAD system used in many areas of the automobile and aerospace industry) into VDDP. Also a CORBA-based interface for the online connection to PDM systems was realized. The crucial property of VDDP, however, is the fact that also users without special VR knowledge can now generate digital prototypes in a considerably shorter time than before. For complex components the effort could thus be reduced from several working days to a few hours.

2000

Augmented Reality – A Mix of Real and Virtual World Supporting Maintenance and Training Julia Mayer

German Abstract Je weiter die Entwicklungen in Richtung digitaler Produktentwicklung und Planung mit Hilfe von VR-Technologien voranschritten, desto klarer wurde, dass die Dokumentation dieser Prozesse in herkömmlicher Form, wie etwa Handbüchern mit Text, Illustrationen und Fotomaterial, nicht optimal war. Ende der 90er Jahre starteten am Fraunhofer IGD erste Versuche, reale Objekte mit virtuellen Daten, Graphiken und Anleitungen zu überlagern und so den Mitarbeitern Unterstützung zu bieten - die Idee der so genannten Erweiterten Realität bzw. Augmented Reality (AR) war geboren und ein erstes Pilotprojekt in Zusammenarbeit mit der BMW AG wurde realisiert. Seitdem war und ist das Fraunhofer IGD in zahlreiche Projekte involviert, die zeigen, wie groß das Potenzial der Augmented Reality für die industrielle Produktion, Service und Wartung sowie für Trainingszwecke von Mitarbeitern ist.

The more the developments proceeded towards digital product development and planning by means of VR technologies the clearer it became that the documentation of these processes in the conventional way, like manuals with text, illustration, and photo material, was not optimal. In the late 1990s first attempts started at Fraunhofer IGD to superimpose real objects with virtual data, graphics, and instructions to give the workers support – the idea of the socalled Augmented Reality (AR) was born and a first pilot project was realized in cooperation with BMW AG. Fraunhofer IGD has since been involved in numerous projects that show how big the potential of Augmented Reality is for the industrial production, service, and maintenance as well as for training purposes.

AVALON - Basis for VR and AR Applications Avalon is an extensible X3d/VRML environment for Virtual and Augmented Reality. It was jointly developed by ZGDV and Fraunhofer IGD. The development started in 1997 and has continued till today. In 2000, the scene-graph OpenSG was included. This resulted in a very powerful system used as a basis for many projects. Using Avalon even computer laypersons will be able to create new applications for different domains by means of script commands. This means that the user can interact with a 3D model previously imported to AVALON, for example the model of a ship, and perform physically based simulations. AVALON enables the use of PC clusters, also the intelligent connection of several computers for

Figure 1: ARTESAS supports service and maintenance workers.

2000

1997 CG

topics 2/2007 13

more computing power, and the use of present-day VR installations like, for example, the CAVE. It can be used under the operating systems Windows and Linux and so be installed on nearly any computer. ARVIKA – Support of Working Processes in Development, Production, and Servicing by Augmented Reality Technologies Based on an idea sketch submitted by Fraunhofer IGD in the scope of a lead project call for tenders in the field of »Human-Technology Interaction« of the BMBF (german federal ministry of education and research) the key project ARVIKA – Augmented Reality for Development, Production, and Servicing – was started in 1999. A total of 23 partners, users from the aircraft and automotive industry and the tool and production machine planning, research and development partners as well as SMEs and integrators - worked more than three years on user-centered application-driven Augmented Reality technologies for an enhancement of working processes in development, production, and servicing for complex technical products. The implementation had three thematic key aspects: ergonomic design of the AR systems, research and development of the AR basic technologies, and integration into the enterprises by pilot application sceneries. ARTESAS – Instrumentation-free Tracking Systems in Servicing Based on the results of ARVIKA Fraunhofer IGD has developed and researched Augmented Reality basic technologies for use in the industrial servicing domain in the BMBF project ARTESAS started in 2003. It was the aim to support service technicians by means of Augmented Reality in the execution of service and maintenance work. The technician is equipped with semi-transparent data glasses and a mobile computer. Via data glasses the AR system gives him a visual presentation of the different work steps directly superimposed to the real field of view. An extensive

Figure 2: within the ULTRA project, Ultra portable Augmented Reality applicatios for industrial maintenance are developed

search for information in printed service and repair instructions is no longer necessary, the enterprise saves time and thus costs. In ARTESAS (»Advanced Augmented Reality Technologies for Industrial Service Applications«) mainly the topics instrumentation-free tracking system, suited for industrial use, and user-oriented AR devices according to technical and ergonomic aspects were treated. Fraunhofer IGD mainly worked on the development of video-based tracking methods and their combination with additional sensor systems. The systems are evaluated in specific scenarios of the application partners. SketchAR – Collaborative Modeling and Visualization in Mixed Realities With SketchAR the researchers of Fraunhofer IGD succeeded for the first time to develop a system that

does not only support a direct modeling in the space but also in the Augmented Reality. With the computer-supported styling system the design phase of the product development can be shifted into Virtual Reality. In contrast to many other VR and AR systems just working with tessellated models SketchAR is based on a CAD kernel providing the whole CAD information including topology and semantics. Data can also be exchanged via interfaces like CATIA, IGES, or SAT, further interfaces are under development. In addition to basic functions like copy, move, and layers SketchAR also offers features to create and modify curves, surfaces, and 3D primitives. Freehand areas of curves are analyzed to bring them into a mathematical form which, if possible, does not distort the original curve progression. For the support in the threedimensional space there are aids like

Figure 3: The Cyberstilo.

1999

1999 14

CG

topics 2/2007

Figure 4: Three dimensional model sketched in SketchAR within a few minutes.

are varnished in blue. The learner wears video glasses replacing all blue shares of the picture visible to him by a virtual scene. For this mix of simulation and reality the blue-box technology is being used, generally known from the film production.

Figure 5: With the Cyberstilo threedimensional sketches iin Virtual Reality can be created.

mirror planes, projection planes, and clip planes, and real 3D snapping and picking. Also the taping technique widely used in the motorcar design to describe characteristic curves at the vehicle, was moved with SketchAR to the virtual space. Furthermore, SketchAR allows collaborative scenarios in which, for example, several users work on a virtual model via their Head Mounted Displays or locally at a physical mock-up. SketchAR is handled with an input tool specially developed for this purpose by Fraunhofer IGD – the Cyberstilo. The Cyberstilo works via a wireless connection and allows freehand sketching of objects in the virtual space. This input tool for product development in the cyberspace has

been developed in collaboration with Barski Design Studio in Frankfurt. VAR Trainer Augmented Reality technologies for use in advanced training have been realized by Fraunhofer IGD scientists with the VAR Trainer (Versatile Augmented Reality Simulator for Training in the Safe Use of Construction Machinery). A combination of mechanics, electronics, automation technologies, computer sciences, and Augmented Reality enables workers to train the handling of different construction machines under realistic conditions. For this aim the learner sits in a driver’s cab. This cab is especially prepared for the Mixed Reality environment. The panes of the cab

2003 2003 CG

topics 2/2007 15

Optimized Simulations Accelerate the Product Development Process Julia Mayer

German Abstract Systeme der Virtuellen Realität und zunehmend auch der Erweiterten Realität sind mittlerweile bei vielen Unternehmen fest in die industrielle Produktentwicklung, Produktion und Wartung integriert. Doch diese Technologien sind hauptsächlich zur anschaulichen Visualisierung statischer Zustände wie etwa des Designs oder bestimmter Wartungsaufgaben geeignet. Ein wichtiger Prozess im Rahmen der Produktentwicklung, die Simulation, kann damit nicht abgedeckt werden. Hier muss noch immer in mehreren Simulationsdurchgängen, quasi durch Versuch und Irrtum, die optimale Lösung gefunden werden. Gerade bei Simulationen komplexer Prozesse ist dies sehr langwierig und kostenintensiv. Die Simulation in den virtuellen Raum zu verlegen und den Optimierungsprozess in bezug auf bestimmte Anforderungen zu automatisieren, würde Industrieunternehmen ermöglichen, neue Produkte innerhalb kürzerer Zeit, in besserer Qualität und preisgünstiger auf den Markt zu bringen. Basierend auf diesen Überlegungen, arbeitet das Fraunhofer IGD seit einigen Jahren an verschiedenen Methoden, um Simulationsprozesse zu automatisieren und so dem Ziel eines vollständig digitalisierten und integrierten Produktentwicklungs- und Produktionsprozesses einen Schritt näher zu kommen.

Virtual Reality and, increasingly, also Augmented Reality systems have become an integral part of the industrial product development process of many enterprises. But these technologies are primarily suited for a clear visualization of static states as for example for design or particular maintenance tasks. They can, however, not cover one important process in the product development: the simulation. Here the optimal solution must still be found in several simulation procedures, which mostly means by trial and error. Especially for simulations of complex processes this is very tedious and cost-intensive. To transfer the simulation into the virtual space and to automate the optimization process in view of particular requirements would enable industrial enterprises to launch new products in a shorter time, in a better quality, and at a lower price. Based on these considerations, Fraunhofer IGD has been working for some years on different methods to automate simulation processes, thus taking another step towards the goal of a completely digitized product development and production process. ViSiCADE – Cooperative Interaction With Simulations Numerical simulations are an important element of the analysis in digital prototyping and, therefore, an essential part of the product development process. The way from an adequate preparation to the representable result of an analysis is, however, time-consuming and error-prone. Cost estimates show that in classic R&D analyses about 90 per cent of the work time is used for data preparation and result evaluation, without in fact considering a modification of the simula-

tions. The scientists of Fraunhofer IGD have for some years now aimed at finding new solutions for a flexible analysis of digital models and prototypes in due course. In the EU project »ViSiCADE – Virtual Environment for the Seamless Integration of CAD/CAE tasks into Virtual Reality« the scientists are working on an innovative simulation environment which, based on VR technology, allows the integration of CAD and CAE in VR. So the time intervals of modeling and evaluation processes are reduced. The VR environment enables to load a CAD model, to derive automatically and interactively an analysis model, and to modify the model. With ViSiCADE different users shall be enabled to interactively influence simulations. Also the entire analysis process is covered, from data editing, identification, and manipulation of the requirements to grid refinement and simulation to post-processing. Since with today’s computational power no real-time analysis is possible with complex models, ViSiCADE will include new methods of »Sub-Modeling«, allowing a simulation-compliant deduction of simpler analysis models. In further development steps the researchers are working on integrating the simulation visualization with AR technologies. By superimposing the visualizations onto the physical prototype different dynamic or static properties of the prototype shall be identified. SR-PRO – Simulated Reality in Product Development With the project SR-PRO the Fraunhofer Institutes IGD, SCAI, and ITWM have developed a working environment for planners, developers, and scientists since 2004. Their concept

2004 2004 16

CG

topics 2/2007

Figure 1: Cooperative analysis of flow field around a turbine (CFD simulation)

Figure 2: Car body model for crashworthiness simulation.

basing on the idea of Simulated Reality. Simulated reality is understood as the approach to combine scientific/technical simulation with methods of Virtual Reality. In the project SR-PRO the researchers are combining applications for current simulation and optimization algorithms with methods from the field of Simulated Reality. They aim at supporting engineers by means of Simulated Reality in a quick search for optimal product solutions and to facilitate decision making. In the scope of the project the scientists are developing an interactive working environment which will allow engineers to visually modify simulation results in real time, to compare different alternatives of a product, and to optimize the design according to the specification of target criteria. A newly created desktop VR work place allows input and interaction

both in 2D and in 3D. The scientists test their vision of Simulated Reality in two application scenarios: the vehicle collision simulation of the car body and the simulation of a production process for nonwoven fabrics. Example crash test simulation: The crash behavior of automobiles has become a crucial purchase criterion. For advertising purposes marketing departments effectively present the classification according to NCAP and the distinction of vehicles with the corresponding stars. For years the automobile industry has banked on the simulation of crash behavior based on digital models to gain insight into the crash behavior at an early stage of the development process. Besides the crash behavior, there are some more, partly competing development goals like the weight reduction and emission minimization so that the develop-

ment process of a vehicle is a multicriteria optimization problem. This is why a user-conducted, interactive optimization process is realized in the project. The user visually explores interim results of the simulation gaining knowledge which will influence the optimization. This example of a crash test simulation furthermore points out another challenge the researchers are facing in the SR-PRO project: If simulations are executed with three-dimensional models where time as a forth dimension plays an important role immense amounts of data accrue. The display of the car model alone can consist of several million elements. To represent the complete deformation of the car in the course of the simulation between 20 and 100 different time steps are stored. So for the complete simulation we can observe data volumes of up to several hundred gigabytes. If the user wants to graphically view and evaluate the results the simulation computer must transfer the data to the local computer. This transfer and archiving often cause great problems in practice. Even latest graphics hardware cannot cope with the data volume and so the transfer via network can cause bottlenecks which again lead to long and unproductive waiting times for the visualization result. To solve this problem the Fraunhofer scientists are using the so-called four-dimensional progressive data transfer in SR-PRO. First a simplified version of the simulation model is sent to the computer. As soon as the system receives further data packages the user receives exact details of the model. This method shows several advantages: As not all data are sent simultaneously there are no transfer restrictions in the network. Also the load for the local computer is low. Furthermore, the system avoids unproductive waiting periods. The model appears on the monitor in a shorter time and so the viewer can start to analyze the simulation while the remaining data are still sent to the computer.

2004

2004 2004

CG

topics 2/2007 17

The New Main Research Areas of Fraunhofer IGD Julia Mayer

German Abstract Eines der entscheidenden heutigen Forschungsbereiche des Fraunhofer IGD ist die Entwicklung von Langzeitarchivierungssystemen für digitale multimediale Modelle. Ziel ist es, dreidimensionale Modelle zukünftig wie Textdokumente archivieren zu können. Damit verbunden ist auch die Suche nach Lösungen für die automatisierte Indexierung der Modelle, Generierung von Abstracts und Inhaltkategorisierung für zukünftige Abfragen und Neugenerierungen aus vorhandenem Material. Ein weiteres zukünftiges Forschungsthema ist der Erhalt von Semantik über den gesamten digitalen Modellierungsprozess hinweg. Das dritte wichtige Feld für die zukünftige Forschung des Instituts ist der Bereich der Überlappung von Computer Vision und Computergraphik. Durch das Zusammenwirken von Computer Vision Technologien und Computergraphik können mehr Anwendungen realisiert werden, die in Unternehmen zur Optimierung von Produktentwicklung, Produktion, Vertrieb, Instandhaltung und Betrieb eingesetzt werden können.

The more the processes related to product development, production, service, and maintenance are digitized and transferred to the virtual space and the more the involved technologies are advanced, the more new questions arise in the research focus of Fraunhofer IGD. So in near future one of the key issues for the industry, and, therefore one of the crucial topics of Fraunhofer IGD today, will be the search for longterm usage and archiving methods for digital multi-dimensional models. This includes the development of solutions for an automatic indexing of the models, for abstract generation, and for a content categorization, since only in this way the models can later be retrieved and new models be generated on the basis of existing material. Another topic that will be in the focus of the Fraunhofer IGD researchers is the maintenance of semantics during the whole digital modeling process. The particular construction processes shall be stored by new ways of model design and the models be provided with additional information to enable for example the described archiving, to accelerate and facilitate the generation of new models, and to facilitate simulations that have been very time-consuming so far. First approaches are offered by the Simulated Reality technologies as already presented. The third major topic the scientists of Fraunhofer IGD around Professor Fellner will deal with is the overlapping area of computer vision and computer graphics. Here new potential for enterprises is developing, for example for a more realistic digital reproduction of products or the use of computer vision technologies for Augmented Reality applications. In the following, the

new key topics of Fraunhofer IGD and their relevance for industrial product development and production processes shall be described in detail. Maintenance of Semantics During the Whole Modeling Process The semantic web, in which information is processed in a way that machines can find it based on its meaning and identify contexts, has been a catchword for some years now. It is indeed relatively easy to define important key terms in text documents. With three-dimensional digital objects like in mechanical engineering it is, however, much more difficult to define and to describe basic properties of a particular object or an object class. But especially in this area the semantic enrichment of models across the modeling process contains a great potential. So new interaction and visualization methods could be realized based on semantic information, the construction process of an object could become comprehensible, and the additional information alone will allow for a reasonable archiving of the digital data. One example for an approach serving to describe three-dimensional forms as well as additional information on the forms is the so-called Generative Modelling Language. Different from traditional modeling languages three-dimensional objects are not described by means of geometric forms but by functions. So in mechanical engineering, for example, not only the final form of the object can be stored but also the construction process itself. Furthermore, the Generative Modeling approach allows to fall back on already solved construction tasks to reuse them in a

2006 2005 18

CG

topics 2/2007

The Generative Modeling Language enriches digital models with semantic information. Graphic: Courtesy of TU Graz, Institute for Computer Graphics and Visualization.

similar situation. In contrast to the conventional representation of threedimensional models in geometric primitives like triangles, points, or NURBS patches, the Generative Modelling Language allows a very compact description of highly complex 3D objects. This description can be evaluated, represented adaptively, and viewed interactively in a very efficient way. Thereby the user can also interactively modify the model parameters. The Generative Modelling can be used wherever digital models are prepared based on firm rules as it is the case for example in the automobile manufacture or in mechanical engineering. The Confluence Between Computer Vision and Computer Graphics Computer vision is already used in different applications in the industry. Computers support for example quality control and measure simple objects. In concurrence of computer vision technologies and computer graphics even more applications can be realized to be used in enterprises for the optimization of product development, production, distribution, maintenance, and service. Based

on computer vision, for instance, in the product design process, photorealistic renderings can be prepared, which give models such a realistic look that they are hardly distinguishable from the realistic model. These renderings facilitate the design reviews and can be used in marketing to offer their customers a realistic impression of the product. Also for Simulated Reality applications, where simulation technologies are combined with Virtual Reality, the images generated by computer vision serve as a basis for the generation of the simulations. On the other hand, the virtual objects faded in in the Augmented Reality are generated with computer graphics technologies, and also for the model-based visualization of objects the computer vision is using models that are generated with computer graphics methods.

CAD and CAE programs and, more and more, also virtual models for visualizing the developed products. Nevertheless, there are hardly any satisfying solutions for a long-term archiving of these models so far. How can sensible abstracts for the digital models be generated? How can they be indexed? How can an automatic content categorization be effected and how can the models be retrieved later, easily and quickly? How can the data be compressed in a way that they need as little storage space respectively as little transmission time as possible? To solve these questions will soon be one of the greatest challenges for the industry in the context of digital product development. Even today, Fraunhofer IGD is working on providing such solutions. It is the aim of the scientists that, in future, it will be able to treat three-dimensional models as it is done today with text documents, which are by default provided with keywords, indexed, archived, and so can be retrieved without any difficulties.

Solution of library-oriented questions in the context of three- and higher-dimensional objects Product development and construction in the industry is increasingly transferred to the digital threedimensional space. Nowadays, we cannot imagine industry without

2006 2007 CG

topics 2/2007 19

20 Years of Computer Graphics – an Interview With Prof. Dr. José L. Encarnação and Prof. Dr. Dieter W. Fellner Dear Prof. Encarnação, dear Prof. Fellner, 20 years of Fraunhofer IGD are an occasion to look back on 20 years of computer graphics. Which were, from your point of view, the most important computer graphics developments in this time? Encarnação: The development of computer graphics during the last decades was less marked by revolutionary leaps than by evolutionary steps. These led then to an »expected« technological revolution about every five years. Nevertheless, we can see significant steps in the development of computer graphics. So one of the early major steps was the creation of a reference architecture for computer graphics which enabled a general program environment. So by creating a reference model with interfaces a general programming level was etablished on which all kinds of applications could be realized. Another important step was the change from vector displays to raster displays. This allowed for the first time to display grey values,

colors, and shadings. They provided a completely new image quality. In addition, the foundation was laid for displaying real 3D objects. As a third important step I see the development of the interaction devices from the simple insert/delete input per keyboard, lightpen, and mouse to the natural interaction as it is used e.g. in the Virtual or Augmented Reality. Fellner: As Professor Encarnação, I also regard the development of new interaction devices as well as the establishment of standards as two of the most important steps in the development of computer graphics. Particularly the latter has, in the first place, made the applications of computer graphics tangible and usable for the industry. In the following years, the progressing of 3D technologies and the acceleration of the graphics hardware were, from my point of view, important points. From the more recent developments, I consider above all (semi-)immersive technologies where the user immerges into virtual worlds, mobile usability

Figure 1: The former Director of the Fraunhofer IGD, Prof. Encarnação (left) and the new Director of the Fraunhofer IGD, Prof. Fellner (right).

up to the »wearability«, meaning the miniaturization of devices up to the integration of computer technology into everyday articles or clothes, and the interaction of computer graphics and computer vision allowing for intelligent environments real milestones of computer graphics. What do you think will be the most important application areas for computer graphics technologies in the near future? Fellner: I see the potential of computer graphics mainly in three domains: First the so-called »closed loops«. It is – as in medicine – mainly about interactively adjusting the planned and designed models to the modified conditions during realization by measurements and imaging systems. As a second important domain I see the continuous use of semantics as it is already aimed at in the field of functional DMU, in which also Fraunhofer IGD is very active. A third important application field is for me the interaction of computer graphics and computer vision especially important for the developing domain of Ambient Intelligence. One of the driving factors I think is the development to »computer graphics on the fly« by increasingly smaller, better, and faster end devices. Encarnação: I agree that Ambient Intelligence is a very important future-oriented domain. But I think that mainly the increasing coalescence of telecommunication, multimedia, and mobility is leading to this development. We all are surrounded by more and more technical appliances. This multitude will, sooner or later, no longer be manageable. That is why there has inevitably to be a paradigm shift. If the devices are as ubiquitous as light and electricity they must also be handled as easily.

2006

2006 20

CG

topics 2/2007

As a second major application area I see the experimental science for example in physics or chemistry. Here user-friendly devices are needed allowing an evaluation of experiments and simulations with real-time feedback to the lab. As a third important area I see the so-called Serious Games transporting the advantages and technologies of computer games to serious, professional application areas like field service, simulation, or planning. The fourth major future application area will be, from my point of view, the digital photography. Currently a whole generation is growing up using digital cameras in their leisure time. To integrate these pictures in private but also professional applications will become an exponentially increasing application area. Prof. Encarnação, you founded Fraunhofer IGD 20 years ago. What was the impetus? Encarnação: At that time the possibilities at university to work in a practice- and application-oriented way were somewhat restricted. Particularly the permanent struggle for moneys for new research projects was cumbersome, protracted, and often affected by university policy reasons. By founding Fraunhofer IGD we could work more application-oriented and computer graphics as a discipline was benefiting from the good reputation of the FraunhoferGesellschaft. Finally the foundation of Fraunhofer IGD was one of the milestones for computer graphics on its way to an established discipline and to an industry-relevant »enabling« technology for the whole IT world. Prof. Fellner, you have long accompanied the institute as a member of the board of trustees, you have now been its director for nearly one year. What were, from your point of view, the most important achievements and accomplishments of the institute in the last years?

Figure 2: Prof. Encarnação (left) and Prof. Fellner (right) during the interview.

Fellner: Besides the technological developments of the institute, it is, from my point of view, the great merit of Fraunhofer IGD that it has considerably contributed to making computer graphics a discipline applicable for the industry and so to establish it in the industry. For a long time, the Fraunhofer IGD was the only research institution to work application-oriented thus making sure that the developments were advanced with the industry and used by it.

tion and communication domain in engineering and construction. For there are so far no tools available allowing to classify the models and animations used in engineering, to identify their content, to generate abstracts or archive them over a long time. We want to realize these things, which go without saying with classic libraries, also for the digital world. Prof. Encarnação, Prof. Fellner many thanks fo the interview.

Where do you see the priorities of the research work of the institute in the next years? Fellner: We will of course continue to use the expertise acquired in the institute and this will allow a future focusing on three major research lines. First this will be the acquisition, the maintenance, and the use of semantics in the whole process chain of modeling. The second focus will be on the overlapping area of computer graphics and computer vision - and will thereby provide the technical basis for the application area of Ambient Intelligence. And, thirdly, we will deal with transferring »library« questions to the informa-

2007

2007 CG

topics 2/2007 21

Go Digital: Toss Pin Boards and Flip Charts Digital Moderation – Experience Active Meetings Dr. Peter Tandler

Even nowadays, the facilitation of workshops and meetings is still typically based on archaic tools: shaky pin boards with flip chart paper, felttip pens, visualization cards and various adhesive or magnetic symbols. Now there is an alternative to these antiquated communication processes and intensive time investments: The Fraunhofer Institute for Computer Graphics Research IGD is now offering Digital Moderation, its marketready digital solution designed in close cooperation with professional facilitators for facilitating business workshops. Digital Moderation improves the effectiveness and efficiency of meetings by combining proven facilitation methods with the innovative possibilities of networked computers. It eases the planning and execution of events, as well as the evaluation, visualization and processing of results. It helps organize and facilitate events, from small work-

shops with 5 participants up to large meetings with 500 people, as well as support the e-moderation of virtual workshops over the Internet. Innovative Companies Need Workshops Workshops are an essential part of operative decision-making processes. Workshops concentrate the expert knowledge of all involved people. Because of this, companies acknowledge the value of workshops for new impulses and for innovation on current and future projects. While the importance and complexity of workshops has constantly been growing during recent years, applied methods and tools for facilitation have basically stayed the same. Inflexible structures like flip charts and glue dots still appear on the agenda. Few participants dominate the discussions, and constructive discussions are limited to group work

German Abstract Digital Moderation steigert die Effektivität und Effizienz einer moderierten Veranstaltung durch die Kombination bewährter Moderationsmethoden mit den innovativen technischen Möglichkeiten vernetzter Computer. Es erleichtert die Planung und Durchführung der Veranstaltung sowie die Auswertung, Visualisierung und Weiterverarbeitung der Ergebnisse. Dabei eignet es sich sowohl für Workshops mit 5 Teilnehmern als auch für bis zu Großveranstaltung mit 500 Teilnehmern, oder zur e-Moderation von virtuellen Workshops über das Internet.

22

CG

topics 2/2007

Figure 1: Contributions are made by participants using laptops. The submitted ideas are collected by a workshop server.

Figure 2: Digital Moderation supports two modes for participants to enter contributions: turn taking – each participant submits a contribution – or group discussion – the group at a table must agree on one common contribution.

and coffee breaks. The documentation of results is cumbersome, time consuming and often even impossible. Advantages of Digital Moderation Digital Moderation brings many advantages. Various methods of facilitation are available, such as single dot questions, rankings and clustering. A special editor allows the quick and easy planning and execution of workshops. The facilitator can observe the whole process at any time. All opinions, contributions, assessments and other activities are processed digitally. The results of a workshop are available immediately. Digital Moderation automatically creates clearly arranged and well-structured meeting minutes for the whole workshop. Superior Efficiency Digital Moderation improves the efficiency of events: All data which is

entered during the meeting is processed digitally and can be presented online immediately. The whole workshop process is more effective and more direct. This is achieved through the parallel input of contributions, a process that makes the time-consuming writing and collecting of results unnecessary. Digital Moderation enhances the facilitator to present the collected data faster and in a better structured way. The time saved can be used for the actual reason for the workshop: finding solutions to problems and creating novel ideas. Improved Participation Parallel input of contributions by participants and the immediate digital processing optimize the data-flow and communication, improving the involvement of your participants. With Digital Moderation, participants perceive group processes as motivating, fast and effective. Results are developed and ranked by all partici-

pants, causing a high degree of identification, motivation, commitment and team-spirit among the group. Even uncommunicative people are motivated, as it is not necessary for them to speak in front of the group or to stick paper cards on flip charts or boards. Higher Quality The motivating dynamics of Digital Moderation have a positive effect on the quality of meeting results. The corporate visualization of results and a better way to solve group problems support a stronger coordination of all participants’ competences. The anonymous submissions reduce the effects of nominal members: there is no way for participants to orient themselves on the contributions of other participants. Thus, pressure to adhere to the hierarchy or to conform to the majority does not occur. This leads to straight and authentic results.

CG

topics 2/2007 23

How high do you estimate the fraction of hepatitis C infected people in different groups?

4%

Blood Donors 2%

Medical Staff

12%

Dialysis Patients

28%

Haemophilia Patients

79%

Drug Addicts 0

10

20

30

40

50

60

70

80

90

100

Figure 3: Participants can rate alternatives numerically, the average results are visualized graphically.

Extensive Scalability: from 5 to 500 Participants Digital Moderation provides flexible scalability. Events, from small workshops with about 5 participants up to large meetings with 500 people, can be organized and facilitated. Better Overview Information technology deals with the counting, evaluating and processing of participants’ contributions. The

consequent use of large projections and screens eases the integration of digital media. Using computer fonts instead of hard-to-read hand writing improves the readability of all ideas and helps structure the contributions. Additionally, the meeting facilitator receives exact feedback from the software regarding the status of the participants’ activities, easily identifying and resolving any issue.

Which psychiatric problems do you expect frequently during a therapy with IFN-alpha

New Facilitation Methods The use of digital information processing provides new ways for the evaluation and presentation of results worked out at meetings. IT support allows for novel problem-solving techniques that are not possible with traditional paper-based workshop facilitation (like economic efficiency calculation or cost-benefit analysis). Tools are developed tailored to the customer’s needs. Processing of Results Digital Moderation provides economic advantages through effectiveness and efficiency: All materials and results are handed over in digital format at the end of the meeting and can be processed directly by the organization’s IT systems, such as office applications or project management tools. Digital Moderation supports the necessary exchange formats such as Excel or XML. Tailored interfaces to the organization´s IT be realized upon request. Flexibility Digital Moderation is flexibly configurable according to many facilitation styles and modularly extensible. The software can be specially customized and adapted to the product according to individual requirements. Link For more information, please visit: www.digital-moderation.com

Points of Contact anxiety disorder

2

paranoid psychoses manias depressions

7

disturbances of memory and concentration fatigue syndrome irritability

Figure 4: Participants can place virtual dots to vote for alternatives.

24

CG

topics 2/2007

2

4

Dr. Peter Tandler Dr. Christoph Hornung Fraunhofer IGD, Darmstadt, Germany E-mail: [email protected] [email protected]

INSCAPE: Interactive Storytelling for Creative People Stefan Göbel, Luca Salvatore, Kristina Cinquegrana

German Abstract Der Bereich Interactive Storytelling – eine Erweiterung der klassischen Geschichtenerzählung um Interaktion – wird zunehmend interessanter für Wirtschaft und Forschung. Der Prozess der Storyerstellung und der Content-generierung, unabhängig von der Art der Geschichte, seien es interaktive Museumsrundgänge, interaktive Schulungen im Bereich Training & Simulation, Edutainment Anwendungen, storybasierte Computerspiele, kann ein langwieriger und komplizierter Vorgang sein. Es existieren zwar bereits einige Autorenwerkzeuge, die den Autor bei diesem Prozess unterstützen, doch besonders bei interaktiven, nichtlinearen Stories, zeigen diese erhebliche Mängel auf. Ziel im Rahmen des integrierten Projekts INSCAPE (EU FP6, IP, IST2004-004150) ist es, dem Autor die Möglichkeit zu geben, seine Ideen auf einfache Art und Weise umzusetzen und daraus spannende interaktive Stories zu erstellen. Die dafür entwickelte Autorenumgebung verzichtet auf textuelle Programmierung und bietet stattdessen Möglichkeiten zur visuellen Strukturierung und Beschreibung der Story.

Introduction Content creation and authoring of Interactive Storytelling based applications, for example interactive theatre, movies, cartoons, video-games, interactive manuals, training simulators, etc., can be a long and difficult process using traditional authoring tools, especially for people without programming skills. Acknowledging the interest of Interactive Storytelling for both authors and users who need to experience more empowering stories, and relying on the increasing relevance of this domain, the aim of the EU funded INSCAPE project (EU FP6, IP, IST-2004004150) is to offer creative people and not only experts, means to readily transform their ideas and mental images into appealing interactive stories using a scalable and intuitive authoring environment. In contrast to most of the existing authoring tools on the market, INSCAPE provides a multi-sensorial (it gets beyond classical screen and mouse paradigms), interactive (it deliberately fosters on creation of interactive stories), multi-participant (it enables development of stories for multiple users at the same time) authoring environment based on a visual language (i.e. the »What You See Is What You Get« principle). INSCAPE breaks the traditional creation pipe where stories are authored and then broadcasted or viewed. All the tools provided by INSCAPE are operable during the experiencing mode of the Story in the INSCAPE system. Hence, the authoring user can partici-

pate in the experience as well and directly modify the scenario according to the user actions and reactions. INSCAPE GUI and Overall Architecture The INSCAPE authoring environment consists of four categories of software components: the core, mandatory plugins, optional plug-ins and standalone applications (see figure 1). The core and the mandatory plug-ins together form the main software application of INSCAPE. The optional plugins share interaction and GUI principles, but appear as separate entities within INSCAPE. The standalone applications are linked and possibly share resources with INSCAPE (for example database, or native formats), but appear as separate applications. In the default setup the core and the mandatory plug-ins are represented in the main GUI with separate windows. These are: – Story Planner – Story Editor – Stage Editor (2D or 3D) and – Object Browser When opening INSCAPE, all of these appear simultaneously, and can be manipulated separately. Figure 2 shows the main INSCAPE GUI in authoring mode: On the left there is the Story Planner. The Story Planner is used as a text editing environment, similar to a basic word processor in which text can by typed or imported/copy and pasted Figure 1: The INSCAPE bundle

CG

topics 2/2007 25

from another source and which contains textual descriptions which are necessary for planning a story. There is also a possibility to import and link reference materials, for example images, sound files, storyboard images, as well as assign expected time for the different sections. The Stage Editor (top-middle) is the primary interaction area where the stage and objects on it are accessed and modified. The Stage Editor also acts as the experiencing window allowing the users to preview their project. Objects with a physical representation (for example characters, props) are visible in actual size, position, etc. Objects with no physical representation (like mouse, sound, scripts) are represented with icons. The Story Editor (bottom-middle) is the place for managing and structuring the story. There are two levels available: the stages and the situations. This is an abstraction of the story. In many domains, a stage refers to the physical surrounding (decor elements) in which the action of the story takes place. In INSCAPE, it refers to a global structure for several situations where common elements (typically the surroundings) can be defined. A situation is a specific phase of the story, defining all surroundings, characters, etc. and their behaviours, similar to a scene in its »cinema« meaning. When selecting a situation in the Story Editor, all objects in that situation are visible in the Stage Editor (for manipulation) and the section of the script in the Story Planner is highlighted. In INSCAPE all assets are treated as objects, once they are imported, whether they are characters, code snippets, sound files, etc. – with distinct representations. The Object Browser (on the right) comes with a set of predefined generic objects (like characters, backdrops, props, cameras, lights). The author can drag and drop the objects into the Stage Editor or into a Stage or Situation in the Story Editor _ both representations are updated either way. A fifth part comes into play when creating an interactive story with INSCAPE: the Object Editor. The Object Editor allows defining all the properties of all objects of the story according to their role and where they belong. This is the visual way to »program« the

26

CG

topics 2/2007

Figure 2: The INSCAPE GUI

interactive story by assigning events and actions to the situations, stages, characters, etc. There are two possibilities to define actions and events: 1. LUA scripts and 2. a PEP-Editor (Plain English Programming) for the non-programming users. The Story Editor in Detail As mentioned in the previous chapter, the Story Editor provides one of INSCAPE’s core functions and authoring environments. The Story Editor bridges relationships between the Story Planner, the Stage Editor and the Objects Browser and articulates basic story structures as well as primary interactive relationships. Functionally the Story Editor splits an interactive story into two components: Stages and Situations. The Stage contains all (often non-interactive) objects which go to make up a particular part of the story and be re-used in several situations. Typical example of this may include backgrounds or scene objects in a 2D or 3D story, ambient sounds, etc. The objects with a physical representation (for example 3D models) can be created with modelling tools like Maya or 3ds Max and can be imported via the Objects Browser. The Story Editor provides a semantic zoom function which shows different levels of detail depending on the zoom level (for example, the objects on the stage are represented as colour coded squares depending on the type or format of the object, icons or a thumbnail preview image within the Stage box). The Stage representation can be set with a background image which may

be a drawing, photograph or piece of text. Furthermore the Stage can be associated with text or images within the Story Planner. A situation contains all the objects within the story that are needed for that particular situation not yet defined at stage level. As with the stage each situation is named and may have an image or piece of text from the Story Planner attached to it and contains iconic representations of objects. Associations/Transitions between stages and/or situations are represented by arrows. Transitions are defined in the Object Editor and would be triggered by some interaction (for example time out, or some interactive event like a click on a hotspot). Apart from the work performed in INSCAPE different additional interfaces have been developed in order to integrate/combine INSCAPE components such as the Story Editor into other systems, for instance as an authoring tool to create and drive games. Current Status and Outlook The INSCAPE project started in 2004 with a 4 year runtime. For the 31st of August 2007 the beta release of the INSCAPE system is planned. People who are interested in evaluating the software will then have the possibility to download a tryout version from the INSCAPE website (www.inscapers.com). Points of Contact Dr. Stefan Göbel, Luca Salvatore, ZGDV Darmstadt E-mail: [email protected] [email protected]

GameDays 2007

Dr. Stefan Göbel, Robert Arthur Konrad

German Abstract Die diesjährigeren GameDays waren dem Thema »Serious Games« gewidmet. In verschiedenen Vorträgen von Experten aus Wissenschaft und Spielindustrie wurden neue Lernanwendungen und Simulationen vorgestellt. Ziel der »Serious Games« es, die motivierenden Aspekte von Computerspielen zu nutzen, um da Interesse zu erzeugen, wo derzeit Langeweile vorherrscht. Ein weiterer Trend, den die »GameDays 2007« zeigten, waren Bewegungsspiele, wie zum Beispiel das virtuelle Tischtennis der TU Chemnitz oder Interaktives Fußball vom ZGDV Darmstadt. Beim abschließenden Familientag konnten Kinder und junge Erwachsene in verschiedenen Kursen eigene Spiele entwickeln. Die Game Days 2008 sind bereits in Planung und werden vom 29. bis 31. Mai 2008 in Darmstadt thematisieren»Educational Games« fokussieren.

From May 31 to June 2, 2007 the third »Game Days« took place in Darmstadt, organized by the Center for Computer Graphics ZGDV e.V., in cooperation with the Fraunhofer Institute for Computer Graphics Research IGD, hessen-it, and the main research areas Computer Graphics and Elearning of TU Darmstadt. Under the motto »Science Meets Business« the series of events acts as intermediary between science and game industry (developers, publishers, and other industries that want to use games for »serious applications«); this year’s focus being on the topic »Serious Games«. Learning applications and simulations are the most prominent examples of these playful applications with a serious background. The motivational aspects of computer games shall be used to create motivation where now boredom is prevailing. The technology for a realistic realtime simulation of complex, threedimensional environments developed in the game industry shall also be used in other domains like training and simulation, marketing and adver-

tising, tourism, or the health sector and sports. On the first two days, a great number of lectures gave professional visitors an insight into different topics around Serious Games. In addition to basic talks, also particular titles and their development were addressed more in detail, for example Global Conflicts: Palestine presented by Dr. Simon Egenfeldt-Nielsen of Serious Games Interactive from Denmark. Furthermore, the lecture also dealt with the problems that may occur during the development of educational games and the current market situation for such games. Sten Hübner of CryTek gave a lecture on the generation of a realistic virtual world using the game Crysis as an example. Although this title is a mere game without any serious pretensions, the explanations of the graphics and physics simulation are also relevant for serious applications, and the game impressively demonstrated the outstanding results of the Frankfurt enterprise in these core areas.

Figure 1: Children developing games.

CG

topics 2/2007 27

A lecture dealing with playful movement games was offered by Prof. Dr. Josef Wiemeyer of TU Darmstadt: Different research projects like the virtual table tennis of TU Chemnitz, interactive football of ZGDV Darmstadt, or commercial game titles and series (for example Wii Sports of Nintendo or EyeToy Kinetics of Sony) show the trend towards active games. »In Germany alone 2607 games came into the market last year. A total turnover of more than one billion euros was achieved«, said Frank Holz, marketing manager of 10tacle, and described that, on the one hand, only a fractional amount of it contains Serious Games (primarily educational games for children); that, on the other hand, however, the trend goes towards this direction and that 10tacle, for example, will increasingly become active in this area. Also politicians support this trend: So, in the context of the opening of the GameDays 2007 in Darmstadt, the government of Hesse has for example offered the »Serious Games Award« endowed with a prize money of 24,000 euros for game developers, explicitly awarding games that can be used for example in advanced training or in the sports and health sector. More information can be found here: www.serious-games-award.de. As in the past years the GameDays were closed with a family day: Besides several hands-on games,

Figure 2: Interactive sudoku (ZGDV)

28

CG

topics 2/2007

Figure 3: Wii Sports in action.

talks, and a »view behind the scenes« the kids and young adults could in different courses develop games of their own. The exceptionally high motivation of the participants and their ultimate success with the development of their own games indicate the highly positive contribution of computer games to the studies of computer science or to the subjects essential for game development like mathematics and physics. The visitors proved such a great interest that some of them asked the organizers for internships (which have already been successfully completed, now resulting in concrete applications for apprenticeships). On all three days, different exhibits could be examined and tested. The ZGDV showed, for example, the learning adventure World of Dinosaurs developed by students. In a comprehensive, three-dimensional museum young people shall fulfill tasks in form of games thus learning something about the dinosaurs. Also exhibited by ZGDV was Interactive Sudoku presented on a screen and, different from keyboard, mouse, and controller, is operated by natural interactions (point gestures and speech commands). The Fraunhofer Institute for Computer Graphics Research and the Games Academy presented with the emo glove a new input device. The glove measures different body values like skin temperature and heart rate. A little dexterity game that becomes the more difficult the more excited the player is demonstrated the functionality.

In two play dens unusual input devices were presented besides classics, curios, and the latest novelties: For instance different movements and music games made the guests sweat; especially the dance mats meanwhile used by different commercial titles proved to be demanding and sweaty. Also presented in the play den was the MyVITness system of OK-Vital. Based on regular home trainers and bicycles (body racers) the system allows to play usual computer games with physical exertion, in this way, highly motivated, working off excess pounds. Nintendo demonstrated with titles like Dr. Kawashimas Brain Jogging and Wii Sports that recently also with the classical game developers there has been a trend towards Serious Games. With such games, the traditional developer and publisher of video games could successfully extend its target group and both achieve a conderable increase in sales and also inspire the visitors of the Game Days 2007. The Game Days 2008 are already in their planning stage and will take place again under the motto »Science Meets Business« in Darmstadt from May 29 to 31, 2008 and will focus on »educational games«. Further information: www.zgdv.de www.zgdv.de/GameDays2007 Point of Contact Dr. Stefan Göbel ZGDV Darmstadt Email: [email protected]

NEWS A-Z

TU Darmstadt Confers Honorary Doctorate upon Hartmut Raffler On January 17, 2007, the Technische Universität Darmstadt conferred an honorary doctorate upon Hartmut Raffler, Director of Siemens AG. The mathematician Raffler received the award »in recognition of his pathbreaking scientific achievements in the field of computer science, both at the university and in industry, and for his visions, his involvement in crucial innovations, and his pioneer work in information and communication technologies and many years of sustainable commitment at the interface between university and industry«. The award was presented by TUD president Professor Dr. JohannDietrich Wörner in the scope of an academic ceremony in the Robert Piloty Building of TU Darmstadt. »Germany owes the work of Hartmut Raffler important innovations in industry and science. He has conse-

Hartmut Raffler, Director of Siemens AG

quently spoken for the dialog between enterprises and universities like the TU Darmstadt. Therefore it is a special honor for us to confer on Hartmut Raffler the degree of Doctor Honoris Causa«, Wörner said. With his work, Dr. Raffler has considerably contributed to establish Germany as an outstanding IT location worldwide. Among the main areas of his work rank, for example service-centered communication, personal agents, intelligent autonomous systems, adaptable human-machine interfaces, IT security, automatic speech recognition, and peer-to-peer networks. Hartmut Raffler was born in Stuttgart in 1946. He studied mathematics and computer science at Technische Hochschule München. After his studies, he worked at Ulm University in the field of stem-cell research, and in 1979 went to Siemens AG, where he became head of the »Software & Engineering« department in 1993. In 1996 Raffler became head of the department »Information & Communications«, comprising about 350 scientists in Munich, Princeton/USA, and Beijing/China. Finally, in 1997, he was appointed director of Siemens AG.

Learning Corporate Success It is a challenge to establish small and medium-sized companies successfully in the long run. To establish and maintain the competing power of a company, the development of promising innovations in the technology and service domain is indispensable. But where do the new ideas come from? Can you learn entrepreneurial creativity? »You can«, Eva-Maria Mahnke says, head of the department Post-professional Education of ZGDV Rostock. »It is important to introduce innovative product development processes purposefully. This implies that connections, strategies, and methods are identified and implemented accordingly. In short, market requirements and company services must be harmonized.« In cooperation with the CIM-Technologie-Zentrum Wismar the ZGDV Rostock offers an on-the-job

advanced training program dealing with »Innovation Management for Products and Processes in Companies«. The qualification started on February 15, 2007 and mainly addresses staff members in the management domain of small and medium-sized enterprises in MecklenburgWestern Pomerania. Two courses with twelve participants each were realized in Rostock and Wismar. The principles of innovation management were worked out and, in a practical part, applied to the specific requirements of the particular company. The special requirements of the enterprises were identified in individual sessions, where the implementation of the learned was accompanied by a coach. Theory, practice, and coaching were always closely related to each other. The project was funded by the European Social Fund and the government of Mecklenburg-Western Pomerania.

IT Companies in MecklenburgWestern Pomerania Think in New Dimensions The new spirit of optimism was almost tangible! Early in January 2007 the partners of the network »Go-3D – Efficient Process Chains for 3D Computer Graphics« initiated by ZGDV had a first meeting. They jointly defined aims and expectations of the future collaboration. The exchange of experiences, the joint preparation of technical solution elements and the commercialization of 3D software products of the network partners are major aspects of the collaboration. Today, 3D models are key applications of computer graphics, used in different businesses. Whether we think of the development of a virtual prototype in car manufacture and shipbuilding or the development of virtual training worlds for the defense of dangerous situations like floods or forest fires – without computer graphics such development were not possible. In Mecklenburg-Western Pomerania, too, significant solutions have been developed in this domain by

CG

topics 2/2007 29

small and medium-sized enterprises. An example are the flexible software and hardware products for the use of Virtual Reality of the Rostock company OTLO VR Systeme GmbH. This and other firms like CiS GmbH and MarineSoft GmbH will collaborate and implement IT solutions via the network for example in the areas of product development, interactive training worlds, and spatial information systems, also including current research results of the scientists of ZGDV, Fraunhofer IGD, or Rostock University. »No doubt, companies and research institutions of the region have developed innovative offers in the 3D domain. But a closer collaboration in research, development, and marketing will certainly increase the success significantly. Especially the nationwide or even international distribution of products of small and medium-sized businesses from Mecklenburg-Western Pomerania is still a challenge«, Dr. Uwe von Lukas said, branch manager of ZGDV Rostock and one of the network managers. »With `Go-3D´ we will bundle the different competences and capacities of the network partners in a way that the different companies will have a sustainable and increasing economic success«, von Lukas continues. Therefore, in the long run, it is the aim to establish the network partners as key experts in Northern Germany and the whole Baltic Sea region beyond the funded run time of three years. »Go-3D« is supported by the program »Netzwerkmanagement-Ost« (NEMO) of the Federal Ministry of Economics (BMWI). The ZGDV, as one of the winners of the selection process was entrusted with the formation and management of this network in Mecklenburg-Western Pomerania. In the next three years the institute will coordinate the collaboration of the participating companies and research institutes.

License for Company Founders More than ninety per cent of all enterprises in Europe have less than ten employees and therefore belong to the so-called micro-businesses.

30

CG

topics 2/2007

Their employees usually have a wide field of responsibility facing challenges as entrepreneur in the enterprise. The education in Europe whether at the university or in vocational schools – does, however, not sufficiently prepare for such challenges. To go into business for oneself is often difficult, because many people lack business management knowledge having also difficulties in assessing what will lie ahead of them. There are sufficient training offers in this area but content and quality are very different and therefore not comparable on an European level. The European project SofE – IDEA School of Entrepreneurs – wants to close this gap: »It is our aim to create a high-quality training standard giving people who are interested in self-employment all over Europe a guideline of which skills and knowledge are necessary for starting their own businesses. At the same time we want to achieve an Europe-wide comparability of knowledge to make it easier for those who have completed starter courses to settle also in other European countries«, Dr. Marion Mienert, manager of INI-Novation GmbH, describes the ambitious objective of the project. »The certification program is graduated into three competence levels. After each level the learner takes a standardized examination, similar to the European Computer Driving Licence ECDL. The result of this test is documented in a certificate. It serves company founders as reference for example to demonstrate investors that they have necessary entrepreneurial knowledge and formation competences«, Mienert explains the certification system. The first level imparts basic knowledge. On this basis increasingly complex competences are transferred. The last level is seen as incentive for a lifelong learning. The issued certificate will lose its validity after some time. Then the learner must acquire new knowledge and prove in another test that he or she has improved her/his skills. The content of the courses covers a total of seven subject areas which are elementary for the work in micro-enterprises. This includes management, marketing

and organization as well as legal and cultural aspects. To keep the content up to date they are regularly reworked by an international expert team. The project has run since October 2005 and will end in September this year. The work has advanced accordingly. »The content of the courses and a model curriculum has already been designed. We are now organizing the examination system for the certificates and defining the franchise concept which will serve to take the exams in any EU country«, Mienert explains the current project status. The project is financed by the European Commission and the European exchange program Leonardo da Vinci.

Venture Capital for iPharro – a Young Enterprise is Taking off It happens again and again that public Internet pages like YouTube unwittingly host big amounts of illegal video footage. As in most cases the metadata necessary for recognition have been removed the tracing of proprietary content is difficult. The iPharro Media GmbH has developed a search machine serving to analyze and identify video clips faster and more efficient than with the human eye. As initial aid the company now has got four million euros from the Fund IV of the Triangle Venture Capital Group Management GmbH. Being the only venture capitalist in Germany focusing exclusively on spin-offs from universities and research institutes, Triangle is now supporting with an investment iPharro, a spin-off of Fraunhofer Institute for Computer Graphics Research IGD in Darmstadt. The software developed by iPharro matches videos in a clip-to-clip process with video information from an existing database. David T. Fisher, manager of iPharro, explains: »According to the US Patent and Trademark Office there are every year losses amounting to 250 billion US dollars due to Internet piracy. Our technology will help to get this problem under control.« This technology, however, is not only useful for the detection of copyright violation, it is also extremely interesting for media research. It

identifies variations in commercials or monitors different TV channels in real time to get an overview of the timing of certain commercials. One of the first customers benefiting from this possibility is Nielsen Media Research. In the scope of a technology screening the technology of iPharro Media GmbH was categorized by the INI-GraphicsNet Stiftung and the ININovation GmbH as marketable. They have since watched and accompanied the utilization consequently, for example by supporting the preparation of a business plan, the identification of financing options, or the search for VC investors. They accompanied the negotiation of the license agreements and the negotiation of the investment agreement. The INIGraphicsNet Stiftung will also continue to act as shareholder and support iPharro in achieving its growth and development goals.

New Life Quality – Technologies for the Generation 65+ The demographic change is one of the greatest challenges for Europe. For a steadily increasing life expectancy of the population also implies that the countries must find new solutions for emerging social and economic problems. Above all the care of old and sick people in residential care homes is a constantly growing expense factor. For the affected persons, too, it is in most cases a great burden to relocate to residential homes, for it implies to give up their personal autonomy. This is where the project PERSONA (PERceptive Spaces prOmoting iNdependent Aging) funded by the EU starts. The vision of the project partners is to provide systems supporting the autonomy, social relationships, and a healthy way of life of older people. They shall be enabled to stay longer in their familiar surroundings and to postpone or even avoid a move in an old people’s home. To realize this vision the researchers are working on a scalable technology platform which shall serve as a basis for future developments in the field of Ambient Intelligence. The platform enables different appliances to network and communicate with each

other. In this way they get reactive environments which are able to analyze the situation of the user, to recognize his preferences and intentions, and to react accordingly. In future, analysis tools could, for example, recognize objects from video data and signal dangers to the occupants. And new, intuitive input tools allow an effective communication and interaction with the environment. Also daily activities like taking a shower, cooking, shopping, keeping appointments, and observing medical instructions shall be supported by the intelligent systems. The idea of this Ambient-Assisted Living (AAL) is not restricted to the living space of the seniors. The technology shall permeate as many areas of life as possible and include as many whereabouts as possible, from the home to the neighborhood to the urban environment. Fraunhofer Institute for Computer Graphics Research IGD is a leading technology partner of the project. The Darmstadt scientists are responsible for the specification of the platform architecture. This means that they see that the platform can be easily adapted to different applications and different devices. They are also responsible for a seamless integration of multimedia and the investigation of novel interaction facilities in AAL environments. In addition, the institute works on the provision of an intelligent middleware. Further information about the PERSONA project can be found on the Internet: www.aal-persona.org.

CityValidator – Quality Control for 3D City Models Three-dimensional city models are more and more used in different applications: so in tourism for the visualization of historical sights, in city planning for the development of noise and disaster control measures, or by providers of navigation devices for the generation of pedestrian and car navigation systems. The models are exchanged in the CityGML format which is based on XML, a standard language developed for the WWW. CityGML allows to consider not only the geometry but also the

semantics and the topology. But testing the model quality means considerable expenses for providers and consumers. If, after the delivery of a model, the customer finds a bug, a troublesome search for the causes will start. Are there problems with the hard- or software of the customer? Or do the delivered data actually contain faults? These questions must be checked, which will take a lot of time and effort. »There is no tool so far able to automatically detect missing elements or syntax errors in the model. The software developers or their customers must either test the data manually or resort to XML validation services which are not very useful for the special requirements of city models,« Jörg Haist, responsible scientist of the department Graphic Information Systems of Fraunhofer IGD, describes the initial situation for the project. With the CityValidator the scientists of Fraunhofer IGD now offer a tool specialized on three-dimensional city models which automatically detects such errors at the push of a button. The CityValidator performs different tests: »The version activated on the Internet checks if the triangulation in the model is correct, if there are holes in the geometries, and if the orientation of the triangles is correct,« Haist explains the method. »In addition we learn if the buildings are textured, a necessary prerequisite to make a city model really intuitive. If all three tests are positive we can assume that the models are correct and complete.« A tedious trouble shooting with the data provider and the client is no longer necessary. To be able to adapt the CityValidator to different applications and requirements the basic version can be extended by semantic tests and included in spatial data infrastructures as Web Processing Service (WPS). The CityValidator is a proprietary development of the Graphic Information Systems department of Fraunhofer IGD and is based on the spatial data server CityServer3D. A first free version is available for download on www.igd.fraunhofer.de/igd-a5.

CG

topics 2/2007 31

Geometric Algebra – Getting Results in a Shorter Time So far, geometric algebra has mainly been used at universities. Now it shall assist enterprises in saving time and costs for the development of technical software. Up to now, several mathematical systems had often to be combined to generate a program in the technical sector. Geometric algebra, however, comprises different systems. So you need just one system which serves to calculate from geometric intuition without having to use pages and pages of formulas. So spheres, circles, or planes become mathematical language tools which can be directly used for computation. In this way, the software development process is considerably simplified and accelerated. In addition, using geometric algebra allows an interactive development. This means that the developers can immediately check the outcome of their work three-dimensionally on the screen and can check from the input if the algorithm works well. »Due to this technology calculations in the field of computer animation can meanwhile be performed up to three times faster. In the future, we expect such a performance also for other applications«, Dr. Dietmar Hildenbrand, scientist at the computer science department of TU Darmstadt, explains. »As many problems from natural science and technology have a geometrical background, geometric algebra lends itself for a great number of applications. We work on making this performance also useful in the field of computer vision and robotics.« On June 11, 2007, Dr. Hildenbrand introduced in his talk »Geometric Algebra and its Application in Robotics« this new technolo-

gy at the University of Applied Sciences Aschaffenburg. He first explained the basic principles of geometric algebra and presented as examples from robotics the inverse kinematics and the grasping process of a simple robot.

Business Knowledge – Efficient, Semantic, Well-structured Often employees are wasting valuable working time going through search results, projects stagnate because important contacts cannot be found in the documents, and expert knowledge remains unused in files which have simply not been found. The unstructured search for information does not only take time it also causes unnecessary costs. The Software ConWeaver puts an end to this. Just one entry and the software browses all the different data sources of a company. ConWeaver does not only include the term entered by the user but also the translation into other languages and the thematic context. Furthermore, the software automatically generates from the company data a semantic knowledge network. The contained data links allow to search for the meaning of a word. So ConWeaver automatically recognizes that the word »client« in the sales database is equivalent to the word »customer« in the e-mail archive and »contractor« in the project documents. »In contrast to conventional search machines ConWeaver creates a relationship between the different data formats. Therefore, the software can efficiently search unstructured as well as structured information sources«, Dr. Thomas Kamps explains, head of the ConWeaver team at Fraunhofer Insti-

Calculation of all angles of a robot at given target positions.

32

CG

topics 2/2007

tute for Computer Graphics Research IGD in Darmstadt. By combining different modules, so-called workflows, the Fraunhofer scientists can adapt ConWeaver to the special requirements of the company. Also Fresenius Medical Care, the worldwide leading provider of services and products for patients with chronic renal failure, wants to benefit from the system. The company has just started to test ConWeaver in the scope of a pilot project in the field of international controlling. At the Science meets Business workshop »Enterprise Search 2.0 – Well-structured and Quickly through the Information Flood« at Fraunhofer Institute for Computer Graphics Research IGD Mr. Kamil Isik, technical project manager with Fresenius Medical Care, got again an overview of present systems for information structuring and search. »The event gave an interesting overview of the different possibilities to structure knowledge in enterprises and to search for information in a quick and carefully targeted way. After a long period of search, we have now found with ConWeaver the system which promises to meet our high requirements. It is especially important for us that the search indexing detects commonalities even in unstructured data and can generate intelligent search lists«, Mr. Isik said. »In the international controlling we have data in many different formats. The search for the right information is often complex and time consuming. The problem of intelligently connecting different data types to collect valuable information for the users has been solved promisingly. I look very much forward to the project as

it will allow us a more efficient use of our data.« You can get further information about ConWeaver on the Internet at: www.conweaver.de.

Fraunhofer IGD Rostock is »Selected Landmark« They puzzle, research, experiment in the land of poets and thinkers. Many good ideas from business and science originate from here and turn the country into a multiple export world champion. But the good idea alone does not suffice. It must also be put into practice! The Fraunhofer Institute for Computer Graphics Research IGD Rostock is one of the think tanks in Germany that makes visions become concrete prototypes. Therefore, the institute has been selected from about 1,500 candidates as one of the 365 landmarks of the national initiative »Germany – Land of Ideas«, presenting this year the Federal Republic of Germany as a modern and creative location for business and science. On June 12, Fraunhofer IGD was awarded a roll of honor distinguishing the house as »Selected Landmark 2007«. The reason for the application was not a matter of mere image cultivation. »Far from it, our concern

was to support the national publicity for Rostock and Mecklenburg-Western Pomerania as IT center of Germany. Especially the rapidly developing market in the Baltic Sea region offers Rostock a unique opportunity to distinguish itself, also internationally, as gateway to the land of IT ideas. We want to benefit from that«, Prof. Dr. Bodo Urban said, head of Fraunhofer IGD Rostock and member of the board of the IT-Initiative Mecklenburg-Western Pomerania. On September 13, 2007 the scientists of Fraunhofer IGD will, therefore, bridge the local IT research and the good local conditions of Rostock. Under the motto »Enjoy IT – Work with the Computer and Relax« they will present jointly with partners from the Rostock University and the ZGDV e.V. (Computer Graphics Center) to the inhabitants of Rostock and surroundings IT developments from the areas of health, tourism, and culture as well as recreation. »We want to show that IT research is not out of touch of the everyday life but is geared to the needs of people to support them in their everyday life«, Prof. Dr. Urban said. Urban thinks of innovations like the EmoHandschuh serving to make the computer recognize the emotions of the user, or the

The Fraunhofer Institute for Computer Graphics Research IGD Rostock was awarded a roll of honor distinguishing the house as »Selected Landmark 2007«.

mobile »DiaTrace« measuring body activities in the daily routine thus encouraging more exercise. The initiative »Germany – Land of Ideas« was created in 2006 on the occasion of the FIFA World Cup by the Federal Government together with the German industry. After the big success of the previous year the series of events is now continued in 2007. In the scope of a competition the organizers were looking again for »365 landmarks in the land of ideas«, one for each day of the year. From more than 1,500 candidates – private and public institutions, cultural and church organisations, memorials, museums, social projects, companies, research centers, or university institutes – an expert jury selected the 365 winners. The project is realized in cooperation with Deutsche Bank.

The Digital Agency: New E-Government Solution for Rostock Time is money! And we were used to lose some of it in the past: due to long waiting lines and processing times at offices. But there will be an end to it for the Rostock citizens. The solution is e-government and promises more citizen-friendliness, transparency, and efficiency in administrative processes. The vision is: In future the citizen will do their administrative paperwork in the virtual office at home. This will save time, money, and in many cases also nerves. In cooperation with the Hanseatic City of Rostock the researchers of ZGDV Rostock have developed in the VESUV project digital assistants that support the citizen in his administrative matters in the network. They also give support for recurring administrative matters like the commercial registration or the organization of big events. »The software allows a digital event management that is unique in this form in Germany«, Guntram Flach says, head of the department »eGovernment & Multimedia Information Management« at ZGDV Rostock. Different German cities like Hannover, Frankfurt am Main, and Görlitz have already indicated their interest in this software.

CG

topics 2/2007 33

»E-government has long come of age and can contribute to dismantle bureaucracy and improve the quality in the administrative work. Therefore, it is time now to make use of the electronic handling of administrative processes area-wide«, Flach continues. The institute is supported by the Computer-Beratungs-Gesellschaft mbH Schwerin, responsible for the industrial development and the distribution of the agent software. A last test run is, however, still missing. Therefore, in the next weeks, the program will be extensively tested in a practice test. In a first phase, there will be an in-house test in the municipal offices, then there will be a user test by the wholesale company Großmarkt GmbH Rostock organizing big events like the HanseSail or the Rostock Christmas market. After the successful completion of these tests the electronic administrative system will also be activated for the Rostock citizens in fall 2007. VESUV is a cooperation project where the advantages of secure software agents that act with legally binding effect are investigated in two concrete applications: e-government und e-tourism. As one of the leading research institutes in Germany for the establishment of electronic information and communication channels the ZGDV is responsible for the project management in the field of egovernment. The scientists of the institute have implemented their ideas and concepts in close collaboration with the municipality, supported by the University of Kassel and the Fraunhofer Institute for Computer Graphics Research IGD in Darmstadt. After a runtime of two years, the final meeting of VESUV took place in Görlitz at the end of June 2007.

With Virtual Ships Heading to Success German shipyards must presently not complain about being short of work. Due to the good international order situation of the maritime industry in this country the turnover has increased by more than four per cent. So German shipbuilders could

34

CG

topics 2/2007

indeed stand up to the European competitors, but the price dumping of the Asiatic competitors continues to put pressure on them. On the long run, they will only be able to extend their competitiveness in the international market by developing technical innovations. An example for the innovation advantage as to the development of new ship types is the use of Virtual Reality technologies. The work with virtual ship models can crucially shorten the entire planning and construction phase and can help in this way to reduce the total development cost. So far, the models and applications could, however, only be produced by IT experts individually and customized. The programming work and above all the capital expenditure was very high accordingly so that potential savings were nullified. »For the future we have to optimize the efficiency of the use of VR technologies. This is the aim of our joint project USE-VR«, Dr. Uwe von Lukas says, head of the Computer Graphics Center ZGDV e.V. in Rostock. »We intend to develop a marketable VR software kit and accompanying services the different customers of the maritime industry can use autonomously and efficiently even if they have no special IT knowledge.« Shipbuilding engineers shall be enabled to generate autonomously virtual models of their projects where they can have a look at every corner of the interior space before production and shape it accordingly to be able to optimize the production. In addition to the ZGDV Rostock and the TU Hamburg-Harburg also the Fraunhofer Institute for Computer Graphics Research and the OTLO VR Systeme GmbH are involved in the research project. Renowned shipbuilders like Aker Yards Deutschland, Meyer Werft Papenburg, Volkswerft Stralsund, and ThyssenKrupp Marine Systems are included on the industry side. They contribute requirements and practice-relevant experience which will be incorporated into the research work of ZGDV Rostock and TU Hamburg-Harburg. It is the aim to extend the supra-regional collaboration of the project partners in the next two years and to transfer it later

to a European level. »USE-VR« is funded by the German Federal Ministry of Research and Technology.

Software Makes Narrowing of Arteries Visible Calcifications in the coronary vessels are the major cause for cardiovascular diseases like cardiac infarctions. They result in different modifications of the vascular walls which lead to a narrowing of the artery, a so-called stenosis. An early diagnosis of such narrowed areas can save life. So far, to detect stenoses, the patient is, in a so-called coronary angiography, injected with a contrast medium through one of the aortas invasively and directly into the heart. In a radiogram this contrast medium makes the heart and the arteries clearly visible to the doctor. For the patient, however, this treatment is a physical strain. CT scan images of the heart have much less impact on the patient, as the contrast medium can be injected into the basilica vein. By improving the imaging technology and newly developed CT scanners enormous improvements of the image quality were achieved in the last years so that they are already used in parts today. The researchers of Fraunhofer Institute for Computer Graphics Research IGD have taken another step forward. Developing the Cardio Analysis Tool they have now generated a software which allows a quick and automatic diagnosis of stenoses on the basis of the CT data. Software Calculates Stenoses Automatically In a first step a specially developed segmenting algorithm extracts the artery from the overall image. For this aim the radiologist only marks the initial and the end point of the artery in the image and another point in the line of the vessel. Then the segmentation is done automatically. In a second step the software measures the extracted artery, compares the local diameter with the environment and can in this way detect irregularities in the run. As analysis

Distinction of the Society of Imaging Informatics in Medicine (SIIM) This essential improvement of the diagnosis possibilities is also acknowledged by the Society for Imaging Informatics in Medicine (SIIM). They awarded Stefan Wesarg and his coworkers M. Fawad Khan of the University Hospital Frankfurt and Evelyn A. Firle of Fraunhofer IGD for their paper with the JDI Best Paper Award 2006, 2nd, in June 2007. Currently, the scientists are looking for industry partners who want to integrate the software into their own systems thus allowing a quicker and gentler diagnosis. At the same time they work on finding additional application areas for the software.

The Cardio Analysis Tool makes narrowing of arteries visible

result he gets a three-dimensional image of the artery where all narrowed areas and also calcifications are marked. »The software developed by us relieves the radiologist of a manual search for variances. Instead all critical areas are indicated and he can then concentrate on these«, Stefan Wesarg, manager of the project at Fraunhofer IGD, describes the advantages of the new method. Above all the sensitivity of the diagnosis is critical. This means that the software must detect really all variances. Clinical Tests Show: High Quality With Double Speed Two clinical tests carried out in cooperation with the University Hospital Frankfurt/Main should show if the software was already suited for the clinic routine. The scientists have analyzed the quality of the system compared to the conventional coronary angiography and compared to the manual analysis of the CT data. »The tests have proved that the new

computer tomography devices deliver images of such a high quality that they can be used without any difficulty for the diagnosis of stenoses. The comparison between the manual and the automatic analysis has also shown that our software detects changes just as well as a health professional. The automatic diagnosis, however, took half the time of the manual examination. It is to the credit of the software that it works absolutely objectively while, for the manual analysis, the subjective perception and the experience of the radiologist play a crucial role«, Wesarg describes the test results. »This implies that the automatic analysis of CT data has two essential advantages compared to the conventional coronary angiography. First of all it is much gentler for the patient as it is non-invasive. And, secondly, the radiologist can notably save time and therefore also cost«, the Fraunhofer scientist sums up. The results of the study were published in the medical magazine »European Radiology« in 2006.

CG

topics 2/2007 35

Events Workshops

Visual Data Analysis – Support for Financial Analysts The economic success of financial institutions is based on precise, quick, and well-founded market analyses and prognoses. For this aim financial analysts must handle increasing amounts of data in less time. The user must find his way in the resulting data jungle to make correct analyses, so he needs software which effectively supports him in his tasks. New chances are offered by the visual analysis, an area of computer graphics. It allows to gain quick findings from big amounts of data. Which graphics applications play a role in financial analysis and how the future of the graphics-supported data analysis could look like was shown in the workshop »Graphical Insight into Financial Data – New Ways through the Data Jungle«, organized by Fraunhofer IGD on January 23, 2007. The Science meets Business workshop of Fraunhofer Institute for Computer Graphics Research IGD also dealt, for example, with the chances and risks of such new developments for banks and insurance companies and the challenges arising from the data processing and evaluation in the financial area. Renowned experts from business and science, including Arnold Wagner (Director Finance of Gartner Deutschland GmbH), Stefan Ott (Reuters), Dr. Björn Fischer (European Central Bank), Gerhard Rauh (Xchanging Transaction Bank GmbH), and Jens Quadbeck (Interactive Data), gave an overview of the IT trends in the finance sector, they spoke about their experiences with a graphics-supported analysis and commented on its impact on different financial analytical working domains. Interesting technology demonstrations and discussions rounded off the event.

36

CG

topics 2/2007

Learntec – Knowledge Worlds Today: Interactive, Dynamic, Cross-linked With conventional means alone, the requirements of today’s teaching and knowledge transfer can no longer be covered. The future belongs to the electronically supported knowledge preparation and exploration. For only improved structures in advanced training and interlinked information and knowledge worlds can help enterprises in Germany to further develop their expertise and the know-how of their staff so that they will be able to keep pace on the international market. The department e-Learning & Knowledge Management of Fraunhofer IGD aims at establishing current research results from the fields of e-learning and knowledge worlds in the application-oriented environment of business and educational institutions. Whether expert management systems or virtual learning worlds, for 15 years now the department has dealt with training and information technologies for the management and transfer of knowledge. At the Learntec fair, February 13 to 15, 2007, the Fraunhofer IGD researchers presented the following projects at the shared Fraunhofer booth: Information – Just in Time! The visualization tool »TM-Viewer« gives enterprises the chance to link fields of knowledge and manage them graphic-interactively. The »TM-Viewer« clearly presents complex issues in form of knowledge maps. It allows an aspect-oriented visualization and can thus be used for example from a manager’s perspective or a co-worker’s point of view. And, in a very short time, the user can extract just the information relevant to his job. Therefore the TM-Viewer is effectively usable in all knowledge-intensive applications like the structuring of learning organizations or the generation of knowledge repositories and so contributes decisively to the efficiency of problem solving. In cooperation with SAP Fraunhofer IGD has developed visualization concepts on the basis of the TM-Viewer technology.

Virtual Learning Worlds With »VAH – Virtuelles Autohaus« (virtual car dealer) the researchers of Fraunhofer IGD have developed a learning platform to supplement the dual education system in the car industry and for advanced vocational training. VAH offers a virtual learning world as bridge between vocational school and enterprise serving as a teaching tool which can also be used as quality assurance tool for the education. »With its 3D learning world character VAH can create and maintain a high motivation of the trainees«, Eicke Godehardt of Fraunhofer IGD explains. »The trainees can for example interact with avatars as their electronic representatives in the world of the virtual car dealer and in this way practice their behavior towards customers. Due to the visual representation of real jobs in the virtual environment the VAH achieves a great practical relevance.« The project is funded by the BMBF (German Federal Ministry of Education and Research). Cooperation partners are the Zentralverband Deutsches Kraftfahrzeuggewerbe e.V. (German Association for Motor Trade and Repair) and the IG Metall (German industrial union). Further information: www.vah-projekt.org DigiMod – Digital Moderation Digital Moderation enhances the effectivity and efficiency of a moderated event by combining proven moderation methods and the innovative technical means of networked computers. Wrapping paper, illegible file cards, and useless photo protocols – the traditional moderation reaches its limits. Digital Moderation is a proven system with computers for moderators and participant groups proving traditional and novel moderation methods. »DigiMod facilitates the planning and realization of the event as well as the evaluation, visualization, and follow-up of the results«, Dr. Peter Tandler of Fraunhofer IGD explains. »So Digital Moderation intensifies the progress of the event, allows more time for the basics, and leads to better results.«

CeBIT 2007 In 2007 too, the scientists of Fraunhofer Institute for Computer Graphics Research IGD presented at the Fraunhofer shared booth some of their latest developments. The more than 500-square-meter sized Fraunhofer booth proved to be a magnet to visitors. BERNIE – Shopping Cleverly Made Easy RFID – the Radio Frequency Identification- allows an automatic identification of objects even across distances. RFID-Chips are for example used for access control, for storing data on tickets, or in the stock and goods management. The researchers of Fraunhofer Institute for Computer Graphics Research IGD have now realized a new scenario: BERNIE, the consultant for nutrition and intelligent shopping. BERNIE is a software serving the user to create his personal nutrition profile. He can for example enter that he is diabetic or vegetarian. During shopping BERNIE compares the ingredients of the goods in the shopping cart with the personal profile of the user and warns him as soon as he chooses a product that does not match his profile. »As a pre-requisite for this scenario’s realization it is necessary that all goods are provided with RFID tags that include the ingredients of the products. By means of these tags the product ‚knows’ itself and informs of its ingredients«, Michael Hellenschmidt explains, responsible scientist at Fraunhofer IGD. »In addition the carts must be equipped with a RFID reader and the buyer must bring his mobile device, for example a PDA or smartphone, with the installed software.« As soon as the user has put the products into the cart the RFID reader reads the data stored on the chips and forwards them to the mobile device. The software then compares the data with the profile of the user. »As the profile is stored on the personal device of the shopper his choice is personalized and, at the same time, absolutely anonymous«, Hellenschmidt explains one of the advantages of the technology. Especially these

properties make the application interesting for supermarkets since it allows them to offer their customers a brand-new service. Prior to a mass application a last decisive hurdle must be taken: »The RFID tags are still too expensive to be widely used in supermarkets«, so Hellenschmidt. If this problem is solved the provision of the products with RFID tags will offer the supermarkets many advantages: So, due to the self-disclosure of the products, stocktaking can be automatized and finished in a matter of minutes. Also the commodities management and the logistics of the products are considerably facilitated by this new technology. In addition, new service offers like BERNIE can be realized. CONFUO©O – Secure, Legal, Peer-2-eer Peer-to-peer (P2P) networks are networks with all connected computers on a par. This means that each computer also serves as server and contributes storage capacity and bandwidth. P2P networks are often used for the exchange of music files or videos. But to many enterprises like disk labels or photography agencies these networks are a thorn in their flesh: Once the content has entered

BERNIE: The consultant software for nutrition and intelligent shopping.

such a P2P network the owners lose any control over its distribution. Track the download of files? Stop the distribution if needed? No chance! On the other side, it is nearly impossible for the user to judge if the exchanged files are proprietary, and if he, by downloading them, becomes liable to prosecution. This is different using the CONFUO©O software developed by Fraunhofer Institute for Computer Graphics Research IGD. »CONFUO©O is as easy to use as any other P2P software. For the rights owner, however, CONFUO©O adds the crucial point that he always retains control of his content«, so Alexander Opel, scientist at Fraunhofer IGD. This means that the rights owner can always retrace who has exchanged his content, and he can, at any time, withdraw his content from CONFUO©O. »Before a content may be exchanged, it must be registered. If a user wants to exchange a song without authorization the exchange is automatically stopped. So providers and users can both be sure that all actions will be legal«, Opel continues. To ensure a high security standard the content and the user registration is outsourced to central servers, so-called »Trusted Third Parties«. These compare the songs according to certain characteristics similar to the melody. In this way it is ascertained that also content that has been compressed or transformed to another data format is recognized. These properties of CONFUO©O allow most diverse business models: »CONFUO©O can, for example, be used by photograph agencies who can then put photographs at the disposal of their customers, or by mobile radio providers enabling their customers to exchange ring tones or music files«, Alexander Nouak, head of the Security Technology department at Fraunhofer IGD, suggests possible business models. 3D Watermarks – Data Under Control Are your digital data safe against theft and fraudulent use? Can you track and control the distribution paths and the identity of the recipi-

CG

topics 2/2007 37

ents of your data? For quite some time now, many enterprises have faced this problem which increasingly also applies to three-dimensional digital models. It is impossible today to imagine things without the clear advantages of an easy and rapid handling and processing of digital data, nevertheless they entail massive security problems: Undesirable manipulation, loss and theft of digital data have become quite usual especially in company networks. One possibility to track the distribution paths of digital data is the integration of digital watermarks. The data are provided with specific additional information so that they can be identified at any time. At the same time, the digital watermark is not perceptible by the user of the data. The watermark can be selected by a secret key which is only known to the author and other authorized users. While research in the field of digital watermarks has so far been limited to image, music, and video data, the Security Technology for Graphic and Communication Systems department of Fraunhofer Institute for Computer Graphics Research IGD has been concentrating for some years on the development of digital watermarks for the tracking and control of the distribution paths of CAD data. The solution developed by Fraunhofer IGD for digital watermarks in CAD models is presently supporting the two major data types in this domain: NURBS and polygon models. High-quality NURBS-based models are generated in the construction and design phase and also used for production, while polygon models are often used in visualization and simulation, recently also for illustrations in PDF documents. The method for NURBS-based models optionally offers a reversible mode: After entering a secret key, an authorized user can remove the digital watermark from the model. This option is especially interesting for production environments in which the digital watermark can be removed from the model of the part just before production, for example in a CNC machine.

38

CG

topics 2/2007

Hannover Fair At this year’s Hannover Fair from April 16 to 20, 2007 the scientists of Fraunhofer Institute for Computer Graphics Research IGD presented several projects around industry and the industrial development and production processes. Application Sharing – rapid and high-quality video transmission No matter if the transferred contents stem from Office packages, CAD applications, or visualization systems, the Realtime Remote Desktop (RRD) continuously records any screen content of the source computer and sends it to the target computer. The data do not have to be modified. So the RRD software is not confined to text-based applications, it also supports all graphics applications which use OpenGL or DirectX. The Realtime Remote Desktop features a transfer rate which is unequaled by any software available on the market. With a comparatively low network load it presents an extraordinary image quality with a high frame rate. This allows for example to play even most complex computer games on remote computers in real time. Furthermore, the Realtime Remote Desktop runs without any difficulty on any customary PC and Microsoft system software and adapts itself to any bandwidth. Mainly visualizationintensive industries benefit from a use of the software, like the automotive domain, the oil and gas industry, the building sector, or the chemical and pharmaceutical industry. A test version of the Realtime Remote Desktop as well as further information on the licensing conditions are offered on the Internet: www.igd.fhg.de/igd-a2/rrds/. From CAD Conversion to High-quality Rendering – Efficiently and Autonomously Usually, conventional CAD converters are only offered in connection with expensive VR processing tools. The Fraunhofer Institute for Computer Graphics Research IGD has therefore developed the conversion software CAD2Vis. For data conversion the user does not need a license

environment for a particular CAD or VR system. For, in contrast to other products available on the market, CAD2Vis enables to convert the data independent of a particular provider. CAD2Vis supports the CAD input formats ACIS SAT, CATIA V4 and V5, IGES, Pro/E, STEP, and VDA-FS. The output can be in the visualization formats OpenInventor 2.1, VRML 2.0, and OpenSG, as well as in the CAD formats ACIS SAT, CATIA V4, IGES, STEP, and VDA-FS. Different from conventional tools CAD2Vis stores the geometries contained in the CAD data also as NURBS model if desired, if this is supported by the target data format. In addition, CAD2Vis offers the possibility to close gaps between adjacent surfaces by »sewing«. The software is, above all, suited for product developers who include VR technologies or Digital Mock-Ups (DMU) into their process chain, looking for an independent solution for the conversion of CAD into VR data. It is also useful for service providers processing VR or web presentations on the basis of CAD data as well as for VR providers offering CAD imports. With the rendering module »SHREg« (Spherical Harmonics Rendering Engine) also developed by Fraunhofer IGD you can make highquality design reviews. The module allows getting a physically based display of the data converted with CAD2Vis in real time. Using the realistic presentation of the models deciders can then for example interactively test different materials or lighting situations, better assess the product quality, and save precious time in search of the best solution. Besides dynamic soft shadows SHREg also supports image-based lighting. Based on an environment exposure the software calculates how an object is illuminated. For this aim SHREg uses the so-called High Dynamic Range Images (HDRI). In contrast to the use of conventional digital photographs the rendering module can correctly record all details of a real scene using HDRI, both in the dark and the light area of a picture. The same method is applied to display flections with a

high volume range. These properties make SHREg especially interesting for Augmented Reality applications. Virtual Simulation – Product Development Support An important step within the process chain is the regular exchange between developers, designers, and customers, because the best possible product variant can only be found together. In the ideal case the different participants can analyze a simulation model and process it interactively in real time from different locations. In practice, however, this cooperative work is often difficult. Due to the transfer of enormous amount of data necessary for example for the simulation of crash tests or streaming simulations the hardware of the computers often reaches its limits. The iFX post-processor of Fraunhofer Institute for Computer Graphics Research IGD sidesteps this problem. With just a minimal need of bandwidth several users at different locations can simultaneously work cooperatively via a network and jointly analyze and evaluate the simulation models. The simulation data, which can sometimes reach the volume of several gigabytes, are forwarded via secure communication channels, for example by DVD. Unlike other postprocessors just control commands and no sensitive data are exchanged via the network between the partners during the cooperative work. For the visualization of the data the researchers rely on advanced graphics hardware that allows to interactively display simulation models with 20 million elements and more. View Into the Virtual World A quicker and more economical product development, a demonstrative transfer of learning content or more realistic, interactive presentation of 3D models – these are some of the advantages the use of Virtual and Augmented Reality offers. Accordingly the two related technologies are increasingly used. Two examples were shown by Fraunhofer Institute for Computer Graphics Research IGD at the Hannover Fair.

The MovableScreen offers an intuitive and interactive access to virtual models.

Using the system MovableScreen the user can intuitively navigate through virtual realities. For this aim, the scientists have installed an interactive display at a rotatable column. Via sensors the rotation of the display is recorded and transferred to the navigation in the virtual world. If the user turns the column this rotation is transferred to the 3D model and he can look around in the virtual model. For the walk through the scenery he presses an interaction puck. A head-tracking camera additionally integrated into the display records the eye movement of the user and orients the display of the virtual scene according to the viewing direction of the user. »Using the MovableScreen even users who have not had any experience with 3D computer graphics so far can navigate through virtual worlds. The adaptation of the display to his line of sight gives the user the impression to look through a window into the virtual world«, so Michael Zöllner, staff member of the department Virtual and Augmented Reality of Fraunhofer IGD. The system can for example be used for a vivid visualization of architecture models. »The MoveableScreen is intuitively operated and usable for the most different application areas, for example for the visualization of virtual factories or plants as well as for the virtual architectural inspection,« Zöllner further explains. The MovableScreen is operated with the VR system Avalon which

has been developed in the scope of the Virtual Competence Network for Virtual and Augmented Reality (ViVERA). In this network five Fraunhofer institutes and their partner universities pool their competences from different sectors to further expand the top position of Germany in the fields of Virtual and Augmented Reality. The research results shall be made available primarily to small and medium-sized companies. Augmented Reality for service and maintenance Together with Rittal GmbH & Co. KG and InnoTeamS GmbH, the scientists of Fraunhofer IGD also presented two mobile AR applications for the support of maintenance workers. In the first application example the service technician gets an AR-supported maintenance instruction. He wears data glasses and has a PDA. With the video camera integrated into the PDA he takes a picture of the machine. Having started the process by speech input the service technician is led step by step through the complex work process. With the data glasses he gets 3D animations which are projected in the correct position onto the machine. These animations explain to him which work must be executed. »The instructions are very intuitive. Even inexperienced workers can carry out complex tasks. In addition, the technician has both hands free for the jobs he has to do and does not have to get along with

CG

topics 2/2007 39

printed manuals«, Dr. Didier Stricker, head of the department Virtual and Augmented Reality at Fraunhofer IGD, describes the advantages of the system. In the second example, the ARsupported teleconsultation, the experienced service expert gives his new colleague support for the maintenance of a machine. The colleague at the machine wears data glasses serving to receive the instructions in form of 3D animations. The pictures of the PDA camera are transferred to the expert in his office. So he can at any time observe the state and the progress of the work. »Of course they can both communicate with each other and the colleague in the office can fade in audio or visual information into the field of view of the service technician on site«, Stricker says. Both developments are based on the results of the EU-IST project ULTRA (Ultra portable Augmented Reality for industrial maintenance applications). Objective of the project is to make use of the increasing performance of mobile end devices like PDA or pocket PCs for AR applications to support mobile service technicians in complex maintenance work. Presently the developments are tested in different companies under real conditions.

VAR-Trainer – Safely Working with Construction Machinery The real-time simulator »VAR-Trainer« (Versatile Augmented Reality Simulator for Training in the Safe Use of Construction Machinery) is a computer-based training tool. Thanks to the combination of mechanics, electronics, automation technologies, computer sciences, and Augmented Reality workers can practice the handling of different construction vehicles under realistic conditions. For this aim the learner sits in a driver’s cab. This cab is especially prepared for the so-called mixed reality environment. This component has been developed by researchers of Fraunhofer Institute for Computer Graphics Research IGD. The panes of the cab are varnished in blue. The learner wears

40

CG

topics 2/2007

In Munich the visitors of the 28th International Trade Fair for Construction Machinery, Building Material Machines, Mining Machines, Construction Vehicles and Construction Equipment bauma could see the VAR-Trainer from April 23 to 29, 2007.

Versatile Augmented Reality Simulator for Training in the safe use of construction machinery

video glasses replacing all blue shares of the picture visible to him by a virtual scene. For this simulation and reality mix the bluebox method is used which is commonly known from film production. So, with this development, Fraunhofer IGD sees that the representation of environment conditions like weather, buildings, or persons is as realistic as possible. The worker can fully immerse into the training scenario. Different from conventional training devices no expensive and unhandy projection systems are needed for the display of the training scenario.

Girls’Day – Computer Science is not a Men’s Domain Gender equality at work is a red-hot topic in Germany. For even though the limits between classical men’s and women’s jobs are much more blurred than some years ago, young women choose disproportionately often »typically female« professions. For years professions like medical assistants or administrative assistants have been at the top of the popularity scale of young girls. But companies increasingly lack junior staff particularly in technical and technology-related areas. Therefore they can and should not do without qualified young women. At the Girls’Day 2007, this year again Fraunhofer Institute for Computer Graphics Research IGD in Darmstadt showed interested girls that information technology is not »for men only«. On April 26, 2007 Fraunhofer IGD opened its doors to 20 young girls at the age of 12 to 15 years. Under the motto »Virtual and Augmented Reality at Fraunhofer IGD« the schoolgirls were informed about one of the most interesting facets of

Girls learned the exciting aspects of computer science.

20 girls visited the Fraunhofer IGD Darmstadt at the Girls’ Day (left).

computer science: computer graphics. In dialog with the scientists of Fraunhofer IGD they could learn all about training facilities and newcomer offers in this domain. Furthermore, lots of live presentations showed the many application areas of computer graphics. A highlight of the event was the demonstration of the CAVE, a projection cube in which the user can directly immerse into the presented virtual scene. The Girls’Day is a nationwide campaign aiming at arousing interest with young girls for occupational and study areas beyond the »typical female« professions. Started in 2001 in close cooperation with the ministry for social affairs and the ministry of education and cultural affairs, the annual future day for young girls became a great success all over Hesse due to a statewide commitment of the cooperation partners.

Clear View in the Information Jungle From the junior to the senior staff to the manager, staff members are increasingly spending their work

time on searching for information necessary to perform their jobs. According to a study of the consulting firm Accenture executives use to invest an average of two hours a day on the information search. Half the data they find are, however, not relevant and thus worthless. This waste of work time involves high cost for the company. The expenses can be even higher if, in the end, the wrong information is used which, according to the Accenture Study, is indeed happening once a week. This can be avoided by using semantic search technologies and company-wide knowledge databases. »New information and communication technologies assist companies in structuring the internal knowledge and make it available to all colleagues«, Dr. Thomas Kamps says, head of the Conweaver development team at Fraunhofer Institute for Computer Graphics Research IGD. The software Conweaver developed by Kamps and his team interlinks the knowledge available in the enterprise and allows a semantic, intelligent, and linked search across all information sources. »With just one query the co-workers can for exam-

ple simultaneously search in the customer database, on the intranet, in the document server, and in the product catalog. Also the translations of the term or related issues are found«, the scientist explains. »As a result Conweaver delivers a clearly structured knowledge network containing for example contact persons, documents, and specifications.« Conweaver and other innovative solutions for a structured and efficient information search in enterprises were introduced by Fraunhofer IGD at the Science meets Business event »Semantics Day: Enterprise Search 2.0 - Well-structured and Quickly through the Information Flood« in Darmstadt on May 22, 2007. Renowned experts from science and business presented new developments, promising companies a way out of the information flood thus helping them to save time and cost. Dr. Uwe Schmid, senior expert of the worldwide Business Technology Office of McKinsey, for example gave in his keynote speech an insight into the challenges of a successful enterprise search, also presenting technical solutions. Joachim Glaubrech of Google Germany presented Google Services as components of the information network. Dr. Jörn Kohlhammer of Fraunhofer IGD showed how new methods of the graphical data analysis qualitatively extend the semantic enterprise search, especially as to quantitative data.

Games Summit - Playing in the Service of the Industry Game technologies and classical industry - at first sight these two areas do not quite seem to fit together. That the industry can strongly benefit from the use of playful applications was shown at the Industry Summit »Serious Games - Playing in the Service of the Industry« on May 30, 2007. In cooperation with the Center for Computer Graphics ZGDV e.V., the Fraunhofer ICT Group organized the Industry Summit in the building of the Hessisches Staatsarchiv Darmstadt. Indus-

CG

topics 2/2007 41

trial representatives and IT experts got to know everything about the applications of Serious Games and could exchange the latest trends in this domain.

Art Meets Science: Recent Photography at the INI-GraphicsNet Stiftung About a dozen artists were presenting their work in the exhibition of the Contemporary Art series in the INI-GraphicsNet Stiftung building. The exhibition started on the occasion of the 3rd Darmstadt Days of Photography, and so the Foundation showed on four floors exclusively photo art. How beautiful, dignified, strong, and yet vulnerable a woman can be, independent of space, age, background, or classic beauty ideal, was revealed by the multiple international awards winning artist Angela Mutti under the title »Women« – in memoriam to Christine 2. Since 1995 Angela Mutti has exhibited her work regularly in individual or shared exhibitions and is publishing in calendars and photo art catalogs. For her photographic work she has for example been awarded several times at the HASSELBLAD AUSTRIAN SUPER CIRCUIT and other international photo competitions. The central issue of her artistic work is the photography of people. »With my pictures I wish to capture people with all their facets. Playing on perspectives, props, forms, and colors, I give my pictures content depth that shall vitalize them and thus emotionally touch the beholders«, the artist said. Klaus Philipp, too, can look back on many years of experience in photography and personal book projects, publications in photo magazines and books, as well as numerous exhibitions. On the premises of the INI-GraphicsNet Stiftung he was showing his picture series »Light Figures«. What reminds of organic forms like amebas, sea animals, or body tissues are in fact reflections of the sunlight on water surfaces resulting from changed projection properties of the photographic lens.

42

CG

topics 2/2007

Uli Gasper has dealt with photography since 1991. In 2003 he started to face up with »LoFi« photography. Here the technology is knowingly brushed the wrong way. Pictures are taken with defective cameras, misexposed and -developed. By a further computer manipulation of the picture blurry, grainy, overexposed pictures are generated recalling the early stage of photography. Besides »LoFi« photography Uli Gasper showed in the INI-GraphicsNet Stiftung under the name of »Lomic« a panel out of four moments that combine with oracular texts comic, photography, and tarot. But also young, less known artists were given a platform by INI-GraphicsNet Stiftung. Three groups of the design department of the University of Applied Sciences Darmstadt presented their work. Under guidance of the photographer and assistant lecturer Andreas Klehm some students dealt in the project »A look back ahead« with classical painters like Rembrandt or Vermeer, whose way to capture mood lighting was unequalled in their time. For the work to be seen in the Stiftung building the students have thoroughly studied, transposed, and photographically reconstructed the work of the Old Masters. The students of Professor Dr. Kris Scholz were restaging in their still lifes »Dead animals, vegetables, bottles, bulky waste« known subjects and thus spanned an array from those days to today. In addition, five students of the design department specializing on photography presented their diploma theses. At the top, in the light-flooded, nearly 5-meter-high staircase of the executive floor of the INI-GraphicsNet Stiftung building, you could find the work of a rather unexpected artist. Here Prof. Dr. Johannes Buchmann presented his »Showcase Faces«. Dr. Buchmann, actually known as professor of computer science and mathematics at TU Darmstadt, shows in his photographs the heads of showcase figures with their faces that appear lascivious, devoted, abstract, cool, or brutal. Taken out of their actual context the scientist stimulates a new approach to the usual.

The only three-dimensional work of art of this exhibition was the object »Memorial angel« of the Weiterstadt artist Dirk Gaydoul. The shining sculpture made of beech wood, brass, high-grade steel, and light-leading Perspex hang in the outer area above the foyer directly in front of the historical prison wall of the former city prison. Intuition of the artist: »Angel memorizes the historical location he is hovering above and bethinks of modern history.«

Accessible Web 2.0 – Access for All?! Web 2.0 has undoubtedly become a buzzword. Stylized to a new hype by many people the subject persists in the media coverage. Web 2.0 is seen as the generic term for different interactive technologies and services of the Internet describing above all a novel user behavior. For the concept of Web 2.0 is based on the principle of »participation«: By providing content himself like for example blogs, the previously passive user changes from a media consumer to a producer. The classical Internet with its static web pages shifts to a dynamic »Social Web«. But how can really all users participate in the new technological and also social developments without excluding certain groups from the content? This was the central question of the 6th congress for an accessible e-government the Center for Computer Graphics ZGDV e.V. organized in Darmstadt on June 20, 2007. At the congress renowned experts showed in their lectures how Web 2.0 can be connected to accessibility. They outlined the new technologies and explained which barriers particular user groups like blind or visually impaired people have to face. At the same time, they showed the risks and chances of an accessible Web 2.0 and presented applications and solutions. In different workshops the participants could, for example, learn how an accessible Web design becomes possible by a correct use of AJAX und Co. or how accessibility can be realized with TYPO3.

10 Years InGeoForum: An Active Platform for a Dynamic Industry Spatial data are used in more and more application areas: Retail businesses use them for an optimal placing of their store, authorities and municipalities use the data for traffic planning purposes and cities realize new, digital travel guides on the basis of this information presenting to their guests the most impressing sights. But to tap the full potential of this dynamic sector it is necessary to create uniform standards and establish a close cooperation between providers and users of spatial data. These are exactly the objectives of the InGeoForum founded in 1997. It was created by Fraunhofer Institute for Computer Graphics Research IGD, TU Darmstadt, Hessen Agentur GmbH, the regional authorities for soil management and geoinformation (Hessisches Landesamt für Bodenmanagement und Geoinformation) as well as the chamber of engineers of the state of Hesse, and serves as platform for information, cooperation, and communication of the parties involved in the geoinformation market, from industry, science, and administration. Meanwhile, nearly fifty members have joined InGeoForum, including renowned enterprises like Dornier, MapInfo, or ESRI, but also many small and medium-sized enterprises. Networked Regionally, Nationally, and Internationally Arising from an initiative of the Rhein-Main region and located in the city of research and science Darmstadt the InGeoForum encourages with its activities the cooperation on a regional, national, and international level. Mainly the many small companies characterizing the sector benefit from this network: »The membership of InGeoForum opens up to SMEs the access to useful information about spatial data and their possible applications«, Dr. Ralf Borchert, spokesman of the management of InGeoForum, explains. Dr. Bodo Bernsdorf, president of DDGI e.V. (Deutscher Dachverband für Geoinformation, German umbrella organization for

geoinformation), agrees: »The geoinformation sector is, in spite of big players like Google-Earth, still strongly characterized by small and medium-sized enterprises. To be able to do an effective lobbying and make their interests heard a close network is necessary. The activities of InGeoForum in the DDGI also contributes to strengthen this network.« Due to the membership of the DDGI in the European umbrella organization for geoinformation EUROGI InGeoForum is also active on a European level. One of the core topics is the Europe-wide harmonization of spatial data aspired by the European INSPIRE initiative. To transform the regulations of INSPIRE, the European framework law for the consolidation and shared use of spatial data in the EU, into national law is one of the major tasks in the next two years. A successful harmonization is also the aim of some InGeoForum members. Under the direction of Fraunhofer IGD they work in the Humboldt project on laying the foundations for the harmonization defined in INSPIRE. But Humboldt is

only one of numerous projects where InGeoForum members are active. Borchert: »The members are active wherever the integration of spatial data from different areas is of interest. This can be the extension of the Frankfurt airport where an incredible amount of data is involved or the EU project Humboldt that shall prepare the ground for a Europe-wide harmonization of geodata.« Anniversary celebrations in the sign of a Europe-wide harmonization Also the anniversary celebrations on the occasion of the 10-year existence of InGeoForum were marked by the Europe-wide integration. In the scope of the conference titled »Geoinformation in Europe« on July 3, 2007 international experts gave an insight into the details of the INSPIRE initiative, presented first results of the HUMBOLDT project and identified possible European funds. The opening talk was given by the newly-elected president of the European umbrella organization for geoinformation, Prof. Mauro Salvimini. He described the European view of the geoinformation sector

From left to right: Dr. Ralf-H. Borchert (spokesman of the InGeoForum management), Prof. Mauro Salvimini (president EUROGI), Ernest McCutcheon, Daniel Holweg (both management InGeoForum), Dr. Bodo Bernsdorf (president DDGI).

CG

topics 2/2007 43

and pointed out the importance of a networking of the European regions also in view of the use of spatial data. Further information about InGeoForum and its members can be found here: www.ingeoforum.de.

Conference among Children: German-French Promotion of Young Talents at Fraunhofer IGD Rostock The promotion of young talents cannot start soon enough: In a virtual video conference preschool children from Rostock and France met recently at the Fraunhofer Institute for Computer Graphics Research IGD. Via Internet they exchanged views with each other in the video lab of Fraunhofer IGD about »Life in the flow«, and so playfully learned the use of the new media. »It is like in the kindergarten«, Christian Peter said, researcher at Fraunhofer IGD Rostock, with a smile. He meant the conference room where children of the AWO day-care center »Rappelkiste« virtually exchanged views with a French peer group from Bois in Northern France. »It is always amazing how quickly kids can communicate with each other in spite of their different mother tongues«, Peter continued, a father himself of one of the day-care kids. The video conference between the kindergartens was organized for the first time last year, then in a small media room of the day-care center. It is since a regular meeting offer for the children, across all country borders. »In the scope of the EU project COMENIUS the children shall be prepared for the new media and, at the same time, develop first foreign-language and intercultural competences«, Marion Hornke, one of the two project coordinators in Rostock, informed. The »Rappelkiste« day care in the Rostock Südstadt is one of the two bilingual day-care centers of the Hanseatic city. Luc Abot, the second project coordinator of the project is teaching the children firsthand French. Like a sponge the little non-native speakers absorbed new words, the Frenchman by birth said.

44

CG

topics 2/2007

Fraunhofer IGD is supporting the virtual globetrotters with the necessary technology on their neighboring institute premises. »It is important to teach children the use of new media early and in a reasonable manner. Therefore we are happy to support computer users and perhaps also computer scientists of tomorrow with technology and expertise«, Prof. Dr. Bodo Urban, director of Fraunhofer IGD Rostock, said. The institute is anyhow a particularly child-friendly employer. »Our colleagues can flexibly organize their work time and also work from home. So it is no problem at all if a child suddenly falls ill and needs her/his parents’ care«, Urban continued.

Summer Academy 2007 – Design Made Easy The Center for Computer Graphics ZGDV e.V. has been internationally known for more than 20 years as specialists for latest developments in the field of computer graphics. So what stands more to reason than also offering courses in graphics programs for designers? Particularly because the application-oriented research institute has been successfully active in advanced training for many years. And so the ZGDV opens its doors for the Summer Academy for designers for the second time this summer with a comprehensive offer of advanced training courses. To adequately allow for the special requirements of designers and people interested in design, the courses are organized in cooperation with the association Hessen Design e.V., last year’s initiator of the Summer Academy. From the design of accessible Web pages to Flash and Adobe courses for all levels to 3D Studio MAX the course offer ranges. To guarantee an intensive, individual support the number of participants is limited to 12 persons per course. The course program, detailed information about the different courses, and registration forms are available on www.zgdv.de.

StudINI INI-GraphicsNet

Researchers and Students at INI-GraphicsNet Due to its international nature, the INI-GraphicsNet is obliged to a long tradition of exchanging researchers and students. Visitors in research and academia from all over the world have been hosted in INI-GraphicsNet institutes, which are adjoined to local universities and participate in university research, teaching and life. The Portuguese Centro de Computação Gráfica (CCG) is related to the University of Minho, CAMTech in Singapore to the Nanyang Technological University (NTU) and imedia, The ICPNM Academy in the US to RISD, the Rhode Island School of Design. The German institutes are adjoined to the Universität Rostock and the Technische Universität Darmstadt. Recently several new institutes joined the INI-GraphicsNet, VICOMTech in San Sebastian/Spain, IGI in Seoul/Korea and GraphiTech in Trento/Italy. And of course not to forget the new partnerships with the affiliated universities. These are the Universidad del País Vasco Euskal Herriko Unibertsitatea (University of the Basque Country), the Ewha Womans University in Korea and Università degli Studi di Trento in Italy. Student exchange programs between IGD and imedia in Providence or CAMTech in Singapore directly support the exchange of students between these institutes. This way is very easy and much less bureaucratic for students to get financial support. But of course there are other possibilities to get funding for exchanges where none of these internal exchange programs apply. Several hints on how to find these scholarships can be found on the studINI Web Site http://www.inigraphics.net/students/ studini/index.html.

Of course the student exchange appointee will assist you too, if you have further questions. Another good starting point for a search for scholarships is http://www.daad.de/. Marie Curie Fellowships for example provide European placements for pre and post-doctoral researchers, usually up to the age of 35, and for experienced researchers. A good place to start searching for Marie Curie Actions, actual proposals and deadlines for applications, is the website http://cordis.europa.eu/mc-opportunities/.

Contact Student Exchange Appointee Fraunhofer Institute for Computer Graphics Research Fraunhoferstrasse 5 64283 Darmstadt Phone: +49 (0) 6151/155-581 Fax: +49 (0) 6151/155-199 E-mail: [email protected] www.inigraphics.net/students/studini

While Marie Curie Fellowships are targeting experienced researchers, there are other funding opportunities for internships. The Leonardo da Vinci program for example supports exchanges for internships within Europe. Due to the increased number of INI-GraphicsNet institutions within the European community, this program seems to be very promising (see also http://europa.eu.int/comm/ education/programmes/leonardo/new/ leonardo2_en.html). Do not hesitate to contact [email protected] for information and for assistance with the application.

ALUMNI Additionally there are some new calls for application for PPP projects, programs for the exchange of persons in predefined projects. These programs are offered by the DAAD and are available for a special exchange country and typically a German project partner. More information on that subject you can find at http://www.daad.de/ppp (in German).

Addressing former staff members of INI-GraphicsNet:

The INI-GraphicsAlumni Forum is a meeting-place and pool for former staff members of the INI-GraphicsNet. If you wish to become a fellow member please contact: Computer Graphics Center Herbert Kuhlmann Rundeturmstrasse 10 64283 Darmstadt Germany Phone: +49 (6151) 155-120 Fax: +49 (6151) 155-450 E-mail: [email protected] URL: alumni.zgdv.de

CG

topics 2/2007 45

GRADUATIONS

Dr. Sascha Schneider November 24, 2006 »Interactive Massive Parallel Visualization of Large Datasets From Fluid Simulations« Supervisors: Prof. Dr. José L. Encarnção Prof. Dr. Reinhard Klein Nowadays, the computer-driven simulation of physical fluid processes has become an important topic. The fields of application are widely diversified, reaching from the area of product development up to safety analysis and optimization. Whereas the simulation in the past was able to benefit from the perpetual parallelization of the available hardware, the respective visualization of fluid datasets still has noticeable deficits. This imbalance results in a situation in which a fluid simulation is able to produce much more

data per time interval than the corresponding visualization is able to handle. To accomplish this fundamental problem of the visualization of fluid datasets, this dissertation builds up several strategic solutions. In this context, the principle of parallelization is transferred to the computation of the visualization. Therefore, a concept is developed to achieve the massive parallel process and display of several concurrent visualization methods, no matter if they are of technical scientific or realistic nature. The parameters of the herein-used visualization methods are interactively steerable, their application on dedicated parts of the data space controllable. Derived from an abstract model of a visualization method and its parallel representation as an object in the system, an important point of the developed approach consists of the encapsulation of individual visualization methods in so called »probes«: Virtual units, whose position, transparency and size can be freely controlled by the user. Individual modules of the system are decoupled through abstract interfaces. This enables an asynchronous and event-driven computa-

Dr. Sascha Schneider relaxedly celebrates his graduation with his doctor adviser Prof. Dr. José L. Encarnção.

46

CG

topics 2/2007

tion of the visualization algorithms, respectively, the geometries produced by them, like particles, volume-rendering or streamlines. Separated from the specialized data formats of the fluid simulation, a progressive data format optimized for the visualization is developed. Using it, a compression of the fluid data can be realized, together with an optimization of the data access for the visualization. Through the massive parallel system and the progressive data format, the scalability of the visualization is achieved. This allows the system to run on different hardware platforms, from laptops to multi-core CPU/GPU systems – always adapted to the available computational performance. Through that, it is possible to display and investigate large fluid datasets even on relatively slow machines and to overcome the imbalance in the performance between fluid simulation and the corresponding visualization.

Dr. Dietmar Hildenbrand December 13, 2006 »Geometric Computing in Computer Graphics and Robotics Using Conformal Geometric Algebra« Supervisors: Prof. Dr. Marc Alexa Prof. Dr. D. Fellner Prof. Dr. W. Straßer In computer graphics and robotics, a lot of different mathematical systems like vector algebra, homogeneous coordinates, quaternions or dual quaternions are used for different applications. Now, it seems that a change of paradigm is lying ahead of us based on conformal geometric algebra unifying all of these different approaches in one mathematical system. Conformal geometric algebra is a very powerful mathematical framework. Due to its geometric intuitiveness, compactness and simplicity, it is easy to develop new algorithms. Algorithms based on conformal geometric algebra lead to enhanced quality, a reduction of development

time, better understandable and better maintainable solutions. Often, a clear structure and greater elegance result in lower runtime performance. However, in this thesis, it could be shown that algorithms based on conformal geometric algebra can be even faster than conventional algorithms. The main contribution of this thesis is the geometrically intuitive and nevertheless - efficient algorithm for a computer animation application, namely, an inverse kinematics algorithm for a virtual character. This algorithm is based on an embedding of quaternions in conformal geometric algebra. For performance reasons, two optimization approaches are used in a way to make the application now three times faster than the conventional solution. These two approaches are based on the code generator Gaigen 2 and on the symbolic calculations with the help of the computer algebra system Maple. They have different advantages: Based on the Gaigen 2 approach, we are able to implement our algorithms in a way that still reflects the

elegant features of Conformal Geometric Algebra. Based on the Maple approach, we are able to implement them with the help of our standard compilers without the need of additional libraries. With these results, geometric computing using conformal geometric algebra will become more and more fruitful in a great variety of applications in computer graphics and robotics as well as beyond these applications in all engineering areas.

Dr. Dietmar Hildenbrand celebrates his graduation.

CG

topics 2/2007 47

Analysis and Interpretation of Quality Metrics for Digital Passport Photos Diploma thesis by: Mathias Frohna Supervisors: ECM. Oriana Yuridia Gonzalez Castillo, Prof. Dr.-Ing. J. Encarnação Background Since the facial image is the mandatory biometric identifier to be included in digital travel documents, the quality of digital passport photos must be guaranteed for the fulfillment of the related international standards, such as the International Civil Aviation Organization Machine Readable Travel Documents specification (ICAO/MRTD 9303) and the International Organization for Standardization (ISO) 19794- 5 »Information Technology - Biometric data interchange formats - part 5 Face image data«. The International Civil Aviation Organization New Technologies Working Group (ICAO/NTWG) recognizes that member states currently and will continue to utilize the facial image as the primary identifier for Machine Readable Travel Documents (MRTDs). Methods to measure the image quality are divided into two categories: objective and subjective. An objective measurement estimates a constant value for the quality of an image through different mathematical algorithms. A subjective measurement estimates a variable value for the quality of an image through different studies, such as direct estimation, visual analog scale or adjetival scales, in which the human opinion of the image plays the main role. Objective The objective of this diploma thesis is to develop an application to integrate and to analyze the quality metrics obtained. The result of the integration and analysis is the subjective description of the quality level of a digital passport photo. Problem description There are three indexes to measure the quality of a passport photo: Image Attributes Quality Index (IAQI), Biometric Content Quality Index (BCQI) and Biometric Content Image Quality Index (BCIQI).

48

CG

topics 2/2007

With this three indexes, a three-dimensional graphic is constructed to show the quality score of the analyzed passport photo; the three indexes are calculated in isolated modules in the same development environment, the results are interpreted manually, and the graphic is produced using other software tools, The whole process consumes too many resources, and there is a dependency of the process expert. Probe strategy The application will be tested using good and bad quality passport photos previously evaluated and qualified with the manual method. Scientific contribution The work specified in this document is a part of a new development to measure the quality of passport photos and to verify that quality can be measured.

Conception and Implementation of a Vegetation Component in a Computer-graphic 3D City Model Diploma thesis by: Beate Geibel Supervisors: Jörg Haist, Prof. Dr. Ralf Dörner, University of Applied Sciences Wiesbaden The concept and implementation of the integration of vegetation into a 3D GIS is explained. The automated representation of plants within a city model introduced in this work serves to simplify the city planning process by a more realistic view and to improve the visual quality of the model. As technological platform the CityServer3D of Fraunhofer IGD is used. Different approaches from the field of vegetation representation are analyzed and tested for suitability for a web-based 3D city model. After this analysis a new approach for integrating and visualizing vegetation objects is elaborated and described.

Adaptive Sampling of Intersectable Models Exploiting Image and Objectspace Coherence Diploma thesis by: Artur Schmidt Supervisor: Anders Adamson In ray casting, interactivity is limited by intersection tests with the underlying surface. Depending on the surface, intersec-

tion tests can be very expensive operations. To achieve interactivity in ray casting in spite of that fact, a framework is implemented that accommodates the amount of intersection tests on the sampling grid in image space according to the surface variation. The shape is approximated by bilinear patches that represent several samples in a rectangular area that can be interpolated, illuminated and tested for occlusion using ordinary graphics hardware, leaving more CPU time for intersection tests. Object-space coherence is exploited by reusing these bilinear patches in consecutive views.

Development of Methods and Concepts for the Creation of a Storytelling-based 3D Action Role-playing Game Diploma thesis by: Thibaut Brusseaux Supervisor: Dr. Stefan Göbel This work presents the development of methods and concepts for the creation of a storytelling-based 3D action role-playing game (action RPG) named Tib’s Quest. Action RPGs have two particular aspects in which we are interested in this work: on the one hand, a real-time action and combat system in a 3D-environment and, on the other hand, storytelling elements, with which the player has to ask the other non-player characters for the next quests to fulfill. We implemented the 3D game engine using the DirectX API in C++. We kept the classic top-down view of the early 2D console action RPGs, which simplifies the rendering optimizations. However, we still need some classic 3D engine features like frustum culling, which we implemented by traversing a quadtree structure. To perform collision detection and response, we implemented a version of a swept sphere algorithm. The terrain structure of the game engine involves classic heightmap and lightmap, and is able to represent both indoor and outdoor environments. The lightmap is computed by ray-casting supported by the quadtree structure. We present and implement a game object component system to represent the game objects populating the virtual world. It is used instead of the usual inheritance-based approach. One advantage of the component-based system is to provide a natural data-driven system

which we use to compose the game objects from outside the game executable. The game objects are geographically registered on a grid to perform efficient proximity tests, such as collisions and interactions. The storytelling relies on the ICML format (Inscape Communication Markup Language), which enables the description of interactive stories, and the Narration Controller, an external DLL, which is able to read ICML stories and manage the story states. The final achievements of this work are the programming of an interface between the Narration Controller and a 3D-game engine as well as the illustrating of the ICML interactive storytelling capabilities in an Action-RPG. We also provide a level editor.

Using Two-dimensional Features to Identify Sheet Music Diploma thesis by: Sonja Becker Supervisor: Dr. Martin Schmucker Nowadays, computer equipment such as copiers and scanners are commonly used devices. By using this equipment, documents, particularly sheet music, can be copied relatively problem-free. The copier can be used to copy music cheaply and quickly. There are few distortions, blots or other markings on the page. The scanner provides quick digital image reproduction, by which the sheet can be sent quickly through the Internet without any further loss in quality. In contrast to conventional copying machines, the copy and the original are identical. For Internet sharing, software called peer-to-peer (P2P) systems are used, which provide networks on which different users can access the files. A well-known example is the exchange system Napster, which facilitated the exchange of MP3 music files over the Internet. Respect for copyrights or for copyright-protected data is rarely taken, which means that illegal copies often result.To detect such misuse, it is important to be able to recognize sheet music clearly. Therefore, properties of sheet music can be used to identify them. For the identification of sheet music, various procedures have already been developed. They can be divided into two classes: watermarking and fingerprinting. Watermarking embeds information (socalled watermarks) into the data, which

creates a modified version of the original. Fingerprinting procedures compute a fingerprint as a characteristic for digital data, which helps to identify the sheet music clearly. They map digital contents to a clear ID, without modifying the data. This thesis evaluates which information of a music sheet can be used to compute fingerprints and which are suitable to clearly distinguish and identify different scores. Therefore, two-dimensional graphical characteristics are used for the identification of sheet music. The characteristics are based on mathematical moments. These moments are statistical factors, which can be used to characterize the sheet music. The two-dimensional pixel distribution of the sheet music on the note sheet is observed. A compact description can be created from the distribution of the pixels. From selected features, a compact representation is tried to be found by using principal component analysis (PCA) and the following quantization. Finally, the result is compared with existing approaches and evaluated.

Conception and Realization of an Interactive Visualization Module for the Configuration and Control of Server Networks Diploma thesis by: Quentin Decavel Supervisor: Thorsten May The goal of this work is to define and implement a module for »Visionapp Control Center«. »Visionapp Control Center« is an application which aims to provide system administrators with a graphical user interface for the automatization of server management inside a network. The module provides an easy way to quickly create server networks, to display information about these networks and to visualize them as completely as possible. Thus, a central part of this work is the development of a so-called »Fisheye«view. Fisheye-views help visualize heterogenous amounts of data by allowing the user to focus on any subset of data while still keeping the whole data set in sight. In our case, the user can choose to focus on a particular connected set of objects and connections inside the network while still being able to see the whole network. This part is essentially a mathematical one, with the characterization of network objects as elements that

move on the surface of a sphere. Another very important aspect of this work is the user-friendliness of the module: the user must be able to create, visualize and expand networks as easily as possible. This demand leads to a need for appropriate user event handling to facilitate the user’s task.

Development of Methods and Concepts for Storytelling-based 2D Adventure Diploma thesis by: Robert Konrad Supervisor: Dr. Stefan Göbel Adventures are an established computer game genre, which, in contrary to other genres, concentrates primary on storytelling. After an introduction about the development of the adventure genre, the thesis deals with the different components constituting the game, which was implemented as part of the thesis. After an overview of the underlying game engine, a discussion of the possibilities of 2D graphics using current computer hardware follows. Then, the story relevant aspects of the game are discussed, which are implemented almost completely in ICML (INSCAPE Communication Markup Language), a language for the description of interactive stories. The final part addresses the implemented concepts, which distinguish the game from other adventures. This is primarily the design of the agent architecture, which is responsible for the behavior of all computer-controlled characters in the game. The appendix includes different implementation details of the game as well as an authoring chapter.

Geospatial Composite Web Services in Three-dimensional Geographic Information Systems Diploma thesis by: Thorsten Itzel Supervisor: Jörg Haist This thesis focuses on geospatial composite web services and their integration into three-dimensional geographic information systems (3D-GIS). The provision of spatial data over commonly accessible web services has enormously increased over the recent years. This success was highly promoted by the standardizing efforts of the non-commercial organization Open Geospatial Consortium (OGC).

CG

topics 2/2007 49

The OGC tries to push the distributed management of geodata even further and is currently developing a web service that enables users to process geodata on distributed resources. A draft specification already exists. The thesis deals with compositions of geospatial web services that provide the possibility to execute complex processing tasks on geodata exclusively on distributed resources. In this paper, current approaches for combining and executing composite web services are considered. Additionally, existing OGC standards and specifications are analyzed for their ease of integration into composite web services. Based on the acquired results, a concept for automated geospatial web service compositions is developed. The concept is evaluated by a prototype implementation of a use case into the CityServer3D.

Automatic Generation of Threedimensional Triangular Meshes from Point Cloud Data Diploma thesis by: Stefan Uhrig, Supervisor: Frank Zeilfelder In recent years, considerable advancements in the development of geometric object scanning technologies were achieved. Typically, so-called point clouds are the result of these scans. They consist of discrete points recorded from the surface of the scanned object. An efficient and high-quality reconstruction of the original object based on the measured data is of high importance in modern computer graphics. Besides the actual surface reconstruction, the conversion to an appropriate non-discrete representation forms a challenging task. The description of surfaces by triangular meshes is still a widely spread standard. This thesis studies both arbitrary topology surface reconstructions from point clouds and the following surface meshing. In particular, modern advancing front algorithms are analyzed. Due to a high level of control on the triangulation, these algorithms are able to produce a high-quality mesh. They reconstruct surfaces by employing a method that projects points in space onto the surface. Providing an efficient and robust projection method therefore constitutes a central requirement of a successful triangulation. Up to now, moving least squares

50

CG

topics 2/2007

were usually used to project points onto surfaces represented by point clouds. We show that practical computation problems occur when using moving least squares in conjunction with advancing front algorithms. These problems are based on the necessity to solve a nonlinear minimization problem. The main contribution of this thesis is the development of a new projection method that is based on Delaunay triangulations. Compared to moving least squares, it features a better reconstruction of surfaces represented by point clouds. Furthermore, our method does not require solving a nonlinear minimization or similar problem, but efficient and robust algorithms for all necessary computations exist. As the direct consequence, the computational costs are lower. Numerous tests resulting from our complex implementations show that these new methods lead to a previously unachieved high robustness and performance in automatic point cloud triangulations.

Generative Programming for Databases Using the Example of MediaSeeker Diploma thesis by: Malte Ried Supervisors: M.Sc. Mareike Behr, René Cavet If many developers work on a project for a database, they often use different ways to access the data. Those different solutions are hard to maintain. Another problem is the changing of the database schema. If a changing is done, big parts of the code have to be changed as well. To solve these problems, one could use a code generator which uses the database schema to produce the code. The presented generator DBCodeGen takes the schema of a running SQLServer database and passes it to one or more output plugins. These are loaded and managed at runtime. The plugin DBCodeExporter is a plugin for DBCodeGen. It produces its output with the database schema and so-called templates. The templates are written in a special language which is interpreted by DBCodeExporter. With this language, it is possible to create loops (like for every table), branches (like if a column is a primary key) and constant data (like the

name of the database). All text, which is not part of the language gets copied to the output. With this technique, it is possible to generate simple text, source code or others which can be written as text. The presented code to access the database is generated with DBCodeGen and DBCodeExporter. If using this code, one cannot easily write a code which does not work. The most possible mistakes were intercepted while compiling the program.

Plan and Control of Operations and Interventions Using Computer Tomography Diploma thesis by: Mira Pommer Supervisor: Christian Dold Medical treatment of bronchial carcinoma (lung cancer) often includes surgical intervention. New challenging bronchial reconstruction techniques (wedge resection, end-to-end anastomosis, y-sleeve) allow the derivation of healthy lung tissue, but require special planning and postoperative supervision. This work presents a semi-automatic analysis of bronchial segments by calculating the run of cross-section diameter and lumen. These measurements are conducted in the original gray-level volume by a combination of the first and second derivatives. Furthermore, there is a bronchialaxis parallel and perpendicular visualization to navigate through the bronchial segment and get a better spacial impression.