Back to IEEE VR 2008 Tutorials

Virtual Reality in the Physical Sciences

Sunday March 9th, 1:30pm - 5:00pm (half-day)
Silver Legacy Resort Casino
Reno, Nevada

Presenters: Oliver Kreylos, Eric Cowgill, Magali Billen

Abstract

Despite tremendous technical advancements since the early HMDs and CAVEs, VR is still not widely accepted for visualization in the physical sciences. VR is viewed by a majority of scientists as a waste of money and effort, "a cute toy," or a "neat way to impress visitors," but not as an effective tool for research. We believe that scientists are not won over by technologies alone, but by tangible results they could not have gotten without using VR. To become mainstream, VR needs to be integrated into the scientific process, as opposed to being used to communicate results afterwards.

In 2003, a group of UC Davis Earth scientists with no prior experience with VR noticed a common urgent need for advanced visualization, and teamed up with computer scientists to form the W.M. Keck Center for Active Visualization in the Earth Sciences (KeckCAVES, http://www.keckcaves.org). KeckCAVES' focus is to develop immersive visualization applications that utilize the full potential of VR to address specific scientific problems faced by the participating scientists. The KeckCAVES experience has shown that scientists will embrace VR if it provides a concrete benefit over standard visualization methods, as long as it is reliable enough to not force them to think about the underlying technology, and easy enough to use that they can focus on their data instead of user interfaces.

This tutorial will relay observations we made while building the hardware and developing the software for KeckCAVES, and show how those approaches that did work can be transferred to other developers aiming to create immersive visualization applications. In short, the approaches we believe are the reason for KeckCAVES' success are:

• Do not replicate everything non-VR visualization does, but focus on what only works in VR
• Focus on users' requirements instead of advancing technologies for their own sake
• Develop simple user interfaces shared by all applications
• Create applications that only do a few specific things, but do them well
• Create applications whose hardware requirements scale with the users' needs

Intended Audience

The tutorial is aimed at VR developers and interested potential users from the physical sciences, and consists of two lectures building on top of one another and two case studies illustrating aspects of interdisciplinary software development:

1. Development of portable VR applications based on the Vrui VR development toolkit (Oliver Kreylos, 1.5 hours)
2. Exploiting VR's strengths for novel approaches to solving problems in the physical sciences (Oliver Kreylos, 1.5 hours)
3. Case study: Use of VR in neotectonic research: Visualization and analysis of 3D point cloud data (Eric Cowgill, 0.5 hours)

Presenters' Background

Oliver Kreylos, kreylos@cs.ucdavis.edu, is an assistant project scientist with the UC Davis Institute for Data Analysis and Visualization (IDAV) and the W. M. Keck Center for Active Visualization in the Earth Sciences (KeckCAVES). Oliver received a Dipl.-Inform. (MS in computer science) from the University of Karlsruhe, Germany, in 1999, and an MS and PhD in computer science from UC Davis in 2001 and 2003, respectively. His main research interest is in interactive and immersive visualization.

Eric Cowgill, cowgill@geology.ucdavis.edu, is an Assistant Professor in the Department of Geology at the University of California, Davis. He has a BA degree from Carleton College (1991), an MS degree from the University of Washington (1994), and a PhD from the University of California, Los Angeles (2001). He started at UC Davis in the fall of 2003 following a 2-year postdoc at the California Institute of Technology. Dr. Cowgill's research seeks to quantify the magnitudes and rates of deformation along major (500-1000km long) intracontinental fault systems to understand their geometric and kinematic evolution, and thus gain insight into continental tectonism.

Magali Billen, billen@geology.ucdavis.edu, is a geophysicist studying plate tectonics, mantle convection, and rock mechanics. She is an assistant professor in the Department of Geology at UC Davis and a collaborating scientist in W. M. Keck Center for Active Visualization in the Earth Sciences (KeckCAVES). Magali received her M.S. and PhD in geophysics from California Institute of Technology in 1998 and 2001, respectively, and has been at UC Davis since 2003. Her main interest is in integrating complex surface observations of deformation and 3D numerical simulation of mantle convection to determine the physical properties of rocks and driving forces for plate tectonics deep in the Earth's interior.