Immersive Visualization of the Hurricane Isabel Dataset

1
Immersive Visualization of the Hurricane Isabel
Dataset

• Kenny Gruchalla and Jonathan Marbach
• BP Center for Visualization
• University of Colorado at Boulder

2
A First Look
3
BP Center for Visualization

• Research in Visualization and Immersive Virtual
  Reality
• www.bpvizcenter.com

4
The Immersive Drilling Planner (IDP)

• A tool for planning oil and gas wells

5
What does that have to do with the Contest?

• We adapted the Immersive Drilling Planner (IDP)
  to load the Hurricane Isabel Dataset
• Added capabilities for time-varying volume and
  vector data
• Added timestep controls
• Added transfer function editing
• Added data probes
• and we called it AtmosV!

6
The Video

• Available at www.jonmarbach.com/isabel

7
AtmosV and the Contest

• Contest Goals
• Interactivity
• Exploration
• Fusion
• Our Solution
• Immersive, Real-time visualization (Performance
  does matter!)
• Timestep control, Real-time Text Probes and
  Temperature Probes, Dataset choice, Transfer
  function editing, Vector-plane selection
• Simultaneous visualization of multiple
  characteristics

8
Challenges

• Performance/Memory Tradeoff (SGI Onyx 3800 vs
  Cluster)
• Vector Volume size (Tried using RGBA encoded
  vectors and Vertex Texture Fetch card/drivers
  arent quite ready)
• Vector Volume size (Tradeoff between full data
  integration and full timestep display)
• Downsampling (Tradeoff between quality and memory
  footprint)
• What does it mean to be Immersed in a Volume?

9
Acknowledgements

• Geoff Dorn, Director, BP Center for Visualization
• Mercury Computer Systems (TGS, OpenInventor)
• NVidia
• The IEEE Vis Contest Committee

10
Contact Information

• Jonathan Marbach
• marbach_at_colorado.edu
• Kenny Gruchalla
• gruchall_at_colorado.edu
• BP Center for Visualization
• University of Colorado
• Campus Box 540
• Boulder, CO 80309
• www.jonmarbach.com/isabel