News | People | Our Github

seeVisible - Parallel Visibility Culling for Large Data Visualization

 



As simulations detailing unprecedented spatial and temporal scales become prevalent in today's scientific research, large data visualization, as a research agenda, has attained a high priority. Herein, one of the primary challenges in large data visualization stems from the forever widening gap between the size of current data sets, commonly amounting to hundreds of gigabytes to terabytes, verus the sustainable system bandwidth available to visualization algorithms. Our past and current efforts on the subject of large data visualization include both algorithm research on view-dependent data culling, and systems research on developing parallel distributed systems. We have already published the following papers on the related topics:

'Distributed Data Management for Large Volume Visualization', J. Gao, J. Huang, C. R. Johnson, S. Atchley, J. Kohl, Proc. of IEEE Visualization Conference, Minneapolis, MN, October, 2005.

'Visibility Culling for Time-Varying Volume Rendering Using Temporal Occlusion Coherence', J. Gao, H. Shen, J. Huang, J. Kohl, Proc. of IEEE Visualization Conference, Austin, TX, October, 2004.

'Visibility Culling Using Plenoptic Opacity Functions for Large Data Visualization', J. Gao, J. Huang, H. Shen, J. Kohl, Proc. of IEEE Visualization Conference, Seattle, October, 2003.