| Abstract | In this paper, we address the problem of the efficient visualization of very largeirregular volume datasets. To this aim, we exploit a multiresolution representation
based on a domain decomposition into tetrahedral cells. A new compact data struc-
ture is described which encodes the whole dataset at a virtually continuous range of
different resolutions, with a storage cost six times lower than a standard data struc-
ture for tetrahedral meshes. Such structure supports on-line selective refinement to
focus resolution on areas that the user considers more critical, based on either field
values, or domain location, or opacity of the transfer function. Selective refinement
is used to trade-off between resolution and speed in visualization, according to user
needs and hardware constraints. These features have been implemented in a new
system, called TAn2 (Tetrahedra Analyzer), for the interactive visualization of three-
dimensional scalar fields defined on very large tetrahedral meshes. Multiresolution
representation and selective refinement make the system fully scalable with respect to
the size of the dataset and to hardware requirements.
|
