vLOD: A Scalable System for Interactive Walkthroughs of Very Large
The problem of interactively rendering and navigating very large
virtual environments has been well-studied in computer graphics.
The two effective methods for efficient rendering of large polygonal
models are to conservatively discard invisible geometry early in the
pipeline and to remove unnecessary detail not discernible from the
current view-point. There have been significant advances recently
in both these areas independently but few comprehensive methods
have been shown to derive maximum advantage of both methods at
the same time. We present vLOD, a scalable system for performing
interactive walkthroughs of very large geometric models on com-modity
graphics hardware. Our system performs work proportional
to the required detail in visible geometry. We use a pre-computation
phase to determine cell based visibility as well as level-of-detail.
This pre-computation generates the geometry visible from a view
cell at the right level of detail. We encode changes between neigh-boring
cellís vLOD, which is not required to be memory resident.
At rendering time, we incrementally re-construct the vLOD for the
current view-cell and render it. Due to the small run-time overhead,
we are able to display models with over tens of million polygons at
interactive frame rates with less than 1 pixel error.
Download Technical Report
Accelerated Walkthrough of Large Spline Models
Download full paper
Interactive Display of Large Solid Models for Walkthroughs.
IEEE Computer Graphics and Applications, March 1996.
Not available online.