Information for media Main
Fact Sheet SIGGRAPH 98
Papers Fact Sheet

SIGGRAPH 98
19-24 July 1998
Orange County Convention Center
Orlando, Florida

SIGGRAPH Papers are the premier annual international forum for the latest and most significant findings in computer graphics and interactive techniques. Papers are submitted for review by a committee of world-renowned experts in computer graphics. At SIGGRAPH 98, 11 paper sessions with a total of 45 papers were presented.

"SIGGRAPH 98 Papers were selected for the significance and potential for future impact of the research, system, or applications they describe," said Michael Cohen, Microsoft Research, SIGGRAPH 98 Papers chair. " The new technologies presented range from low-level graphics hardware to new methods for creating realistic human face models that can be used in a variety of interactive modes."

Highlights

Facial Modeling & Animation
Chair
Irfan Essa, Georgia Institute of Technology

Creating interactive virtual humans is somewhat of a holy grail for computer graphics researchers. The most expressive part of a human, and thus the most difficult aspect of the problem is to create realistic facial models and make them move like a real human. The papers in this session addressed present solutions to much of this problem. This is of great interest for animators, Web designers, game developers, and artists.

Making Faces
Brian Guenter, Cindy Grimm, Henrique Malvar, Microsoft Research
Daniel Wood, University of Washington

A system for capturing both the three-dimensional geometry, and color and shading information for human facial expressions has been designed. Images are used to create a texture map sequence for a three dimensional polygonal face model which can then be rendered on standard 3D graphics hardware. The resulting facial animation is surprisingly life-like and looks very much like the original live performance.

An Anthropometric Face Model Using Variational Techniques
Douglas DeCarlo, Demitri Metaxas, Matthew Stone, University of Pennsylvania

A system that automatically generates varied geometric models of human faces was described. Anthropometric statistics for lively face measurements in populations are used to generate random faces of people who have never existed.

Modeling Realistic Facial Expressions From Photographs
Frederic Pighin, Jamie Hecker, David H. Salesin, University of Washington
Dani Lischinski, The Hebrew University
Richard Szeliski, Microsoft Research

New techniques for creating photorealistic textured 3D facial models from photographs of a human subject, and for creating smooth transitions between different facial expressions were presented.

Image-Based Modeling & Rendering
Chair
Julie Dorsey, Massachusetts Institute of Technology

Image-based modeling represents a new and very exciting intersection between computer vision and computer graphics technologies. Papers in this category showed how researchers are developing new data structures to allow the use of captured imagery beyond simply mapping the images onto the surface. In particular, images can be warped and blended to create an illusion of being immersed in a full three dimensional space.

The Office of the Future: A Unified Approach to Image-Based Modeling and Spatially Immersive Displays
Matt Cutts, Henry Fuchs, Adam Lake, Ramesh Raskar, Lev Stesin, Greg Welch, University of North Carolina at Chapel Hill

Ideas, proposed technologies, and initial results for an office of the future based on a unified application of computer vision and computer graphics were discussed. In particular, through the use of cameras and projection methods, virtual portals between distant offices can be created. An initial implementation of the overall vision, and preliminary demonstrations of the dynamic modeling and display techniques were shown.

Rendering Synthetic Objects Into Real Scenes: Bridging Traditional and Image-Based Graphics With Global Illumination and High Dynamic Range Photography
Paul Debevec University of California, Berkeley

A method that uses measured scene radiance and global illumination algorithms to realistically render computer-generated objects into real-world scenes with correct lighting was presented. Applications for the method exist for visual effects, interior design, and architectural simulation.

Plants, Palettes, Perception
Chair
Kellogg Booth, The University of British Columbia

The papers in this session covered a variety of topics ranging from creation and display of large ecosystems of plants, to how to tune computer graphics output to the specific perceptual aspects of human vision.

Realistic Modeling and Rendering of Plant Ecosystems
Oliver Deussen, The University of Magdeburg
Patrick Hanrahan, Matt Pharr, Stanford University
Bernd Lintermann, The ZKM Center for Art and Media, Karlsruhe Radomir Mech, Przemyslaw Prusinkiewicz, The University of Calgary

Modeling and rendering realistic natural scenes consisting of thousands of individual plants to produce realistic visual effects and complete computer graphics environments is particularly challenging. The system presented shows how to generate complex and visually rich scenes of plant ecosystems consisting of more than 50 million primates.

Other papers sessions:

Animation & Simulation
Multiresolution Surfaces
Hardware Acceleration
Image-Based Rendering
Rendering
3D Interaction
Surfaces
Art, Illustration, Expression

Main Comments and questions