EFFICIENT VISIBILITY FOR DISTRIBUTED VIRTUAL URBAN ENVIRONMENTS

This book focuses on the utilization of occlusion
culling for the real-time visualization of
distributed virtual urban environments. Today's
graphics hardware renders all the primitives in any
order and uses z-buffer to determine which primitives
are visible on a per-pixel basis. However, visibility
is computed in the last stage of rendering pipeline
and every rendered primitive is not visible in the
final image. Early culling of the invisible
primitives in a complex scene is valuable for
efficiency in the conventional rendering pipeline.
This may reduce the number of primitives that will be
processed in the rest of the pipeline. In this work,
we propose an efficient visibility estimation method
for distributed virtual urban environments. The
proposed method is based on occlusion culling to
identify and cull the occluded parts of the scene.
This not only computes conservative potential visible
set (PVS) for each client but also assures the
computed PVS to be available at the client on-time
and reduces the network traffic by grouping the
clients which may see each others dynamically.

Autorentext

Gürkan KOLDAS is graduated from Turkish Naval Academy in 1993. Hehad MS and PhD degrees from the Middle East Technical University,Turkey. He is still Lt.Cdr. in Turkish Navy and working as achief of system administration. His researches focus on dynamicsimulation, visibility, real-time rendering and distributedsimulation. Veysi Isler is a faculty member at the Department of Computer Engineering, Middle East Technical University, Turkey, where he is also the director of Modelling and Simulation Research and Development Center. His research interests include rendering, visualization, game technology, and parallel rendering algorithms. Isler has a Ph.D. degree in Computer Engineering in paralel rendering. Contact him at veysi@metu.edu.tr

Klappentext

This book focuses on the utilization of occlusionculling for the real-time visualization ofdistributed virtual urban environments. Today'sgraphics hardware renders all the primitives in anyorder and uses z-buffer to determine which primitivesare visible on a per-pixel basis. However, visibilityis computed in the last stage of rendering pipelineand every rendered primitive is not visible in thefinal image. Early culling of the invisibleprimitives in a complex scene is valuable forefficiency in the conventional rendering pipeline.This may reduce the number of primitives that will beprocessed in the rest of the pipeline. In this work,we propose an efficient visibility estimation methodfor distributed virtual urban environments. Theproposed method is based on occlusion culling toidentify and cull the occluded parts of the scene.This not only computes conservative potential visibleset (PVS) for each client but also assures thecomputed PVS to be available at the client on-timeand reduces the network traffic by grouping theclients which may see each others dynamically.