Peter Wißkirchen – författare

Die Computergraphik kann anspruchsvolle Visualisierungsaufgaben nur mit Hilfe wissensbasierter Methoden sinnvoll beherrschen. Umgekehrt sind KI-Systeme mit wachsender Komplexität auf graphische Visualisierungshilfen angewiesen, um Modelle und Zustände überschaubar zu machen. Dies bewirkt eine zunehmende Verflechtung der Disziplinen Graphische Datenverarbeitung und Künstliche Intelligenz. Der Tagungsband enthält 11 Vorträge des GI-Fachgesprächs Graphik und KI. Die Beiträge zeigen das Zusammenspiel von Computergraphik und KI beim Entwickeln von intelligenten graphischen Oberflächen, beim wissensbasierten Präsentieren von Informationen, beim automatischen Generieren von Graphiken und in Anwendungen aus den Bereichen CAD, großtechnische Systeme und Robotik. Das Buch gibt damit einen Überblick über wichtige Felder, in denen Graphik und KI integriert eingesetzt werden, und zeigt den Stand der Entwicklung. Es richtet sich sowohl an Forscher und Entwickler in den Bereichen Graphik und KI als auch an potentielle Anwender intelligenter Graphiksysteme und benutzerfreundlicher KI-Systeme.

The area of computer graphics is characterized by rapid evolution. New techniques in hardware and software developments, e. g. , new rendering methods, have led to new ap­ plications and broader acceptance of graphics in fields such as scientific visualization, multi-media applications, computer aided design, and virtual reality systems. The evolving functionality and the growing complexity of graphics algorithms and sys­ tems make it more difficult for the application programmer to take full advantage of these systems. Conventional programming methods are no longer suited to manage the increasing complexity, so new programming paradigms and system architectures are re­ quired. One important step in this direction is the introduction and use of object-oriented methods. Intuition teils us that visible graphical entities are objects, and experience has indeed shown that object-oriented software techniques are quite useful for graphics. The expressiveness of object-oriented languages compared to pure procedurallanguages gives the graphics application programmer much better support when transforming his mental intentions into computer code. Moreover, object-oriented software development is a, weil founded technology, allowing software to be built from reusable and extensible compo­ nents. This book contains selected, reviewed and thoroughly revised vers ions of papers submit­ ted to and presented at the Fourth Eurographies Workshops on Object-Oriented Graphics, held on May 9-11, 1994 in Sintra, Portugal.

The area of computer graphics is characterized by rapid evolution. New techniques in hardware and software developments, e. g. , new rendering methods, have led to new ap­ plications and broader acceptance of graphics in fields such as scientific visualization, multi-media applications, computer aided design, and virtual reality systems. The evolving functionality and the growing complexity of graphics algorithms and sys­ tems make it more difficult for the application programmer to take full advantage of these systems. Conventional programming methods are no longer suited to manage the increasing complexity, so new programming paradigms and system architectures are re­ quired. One important step in this direction is the introduction and use of object-oriented methods. Intuition teils us that visible graphical entities are objects, and experience has indeed shown that object-oriented software techniques are quite useful for graphics. The expressiveness of object-oriented languages compared to pure procedurallanguages gives the graphics application programmer much better support when transforming his mental intentions into computer code. Moreover, object-oriented software development is a, weil founded technology, allowing software to be built from reusable and extensible compo­ nents. This book contains selected, reviewed and thoroughly revised vers ions of papers submit­ ted to and presented at the Fourth Eurographies Workshops on Object-Oriented Graphics, held on May 9-11, 1994 in Sintra, Portugal.

Object-oriented systems have gained a great deal of popularity recently and their application to graphics has been very successful. This book documents a number of recent advances and indicates numerous areas of current research. The purpose of the book is: - to demonstrate the extraordinary practical utility of object-oriented methods in computer graphics (including user interfaces, image synthesis, CAD), - to examine outstanding research issues in the field of object-oriented graphics, and in particular to investi- gate extensions and shortcomings of the methodology when applied to computer graphics. Papers included in the book extend existing object-oriented graphical techniques, such as Smalltalk''s "model view controller" or "constraints", introduce the use of complex and persistent objects in graphics, and give approaches to direct manipulation interfaces. The reader is presented with an in-depth treatment of a number of significant existing graphics systems, both for user interfaces and for image synthesis. There are theoretical surveys and chapters pointing to new directions in the broad field of computer graphics. Computer language scientists will find a useful critique of object-oriented language constructs and suggested ways to extend object-oriented theory.

Object-oriented systems have gained a great deal of popularity recently and their application to graphics has been very successful. This book documents a number of recent advances and indicates numerous areas of current research. The purpose of the book is: - to demonstrate the extraordinary practical utility of object-oriented methods in computer graphics (including user interfaces, image synthesis, CAD), - to examine outstanding research issues in the field of object-oriented graphics, and in particular to investi- gate extensions and shortcomings of the methodology when applied to computer graphics. Papers included in the book extend existing object-oriented graphical techniques, such as Smalltalk's "model view controller" or "constraints", introduce the use of complex and persistent objects in graphics, and give approaches to direct manipulation interfaces. The reader is presented with an in-depth treatment of a number of significant existing graphics systems, both for user interfaces and for image synthesis. There are theoretical surveys and chapters pointing to new directions in the broad field of computer graphics. Computer language scientists will find a useful critique of object-oriented language constructs and suggested ways to extend object-oriented theory.

At present, object-oriented programming is emerging from the research labora­ tories and invading into the field of industrial applications. More and more products have been implemented with the aid of object-oriented programming techniques and tools, usually as extensions of traditional languages in hybrid development systems. Some of the better known examples are OSF-Motif, News, Objective-C on the NeXT computer, the C extension C++, and CLOS an object­ oriented extension of LISP. All of these developments incorporate interactive graphics. Effective object-oriented systems in combination with a graphics kernel­ does it mean that the field of computer graphics has now become merely an aspect of the object-oriented world? We do not think so. In spite of interesting individual developments, there are still no sound object-oriented graphics sys­ tems available. If it is desired to develop a complex graphics application embed­ ded in a window-oriented system then it is still necessary to work with elemen­ tary tools. What is to be displayed and interactively modified inside a window must be specified with a set of graphics primitives at a low level, or has to be written with a standardized graphics kernel system such as GKS or PHIGS, i. e. , by kernels specified and implemented in a non-object-oriented style. With the terms GKS and PHIGS we enter the world of international graphics standards. GKS and PHIGS constitute systems, not mere collections of graphics primitives.