Techniques for the multiresolution management of volume datasets have been devised.
in TAn v.1
The first release of the TAn
system supports the multiresolution management of tetrahedra meshes on top
of a Delaunay-based refinement approach. The first approach
supported a small number of different representations, organized into a pyramid-like structure, where each level i gives an
approximation of the represented scalar field with an error which does not exceed a
in TAn v.2
We have developed a new compact data structure for efficiently encoding the whole dataset at a virtually continuous range of different resolutions. Such structure supports the efficient extraction of simplified meshes that approximate the original dataset at a resolution depending on either a spatially-defined filter or depending on critical field-values. The mesh to be rendered can be selectively refined over areas that the user considers more critical, on the basis of either field values or domain locations. Extraction criteria are determined based on the current viewing modality and on the physical limitations of the system.
The corresponding paper is currently under revision.
Multiresolution image synthesis algorithms were also studied, in order to allow the simultaneous use of multiple resolution levels while rendering a single image.
A first approach proposed called MagicSphere, implemented the idea of a "focus area" in the visualization of isosurfaces fitted over a multiresolution model. To avoid to cope with the "cracks" that are produced when isosurfaces are fitted on a multiresolution data representation, MagicSphere was designed as a generic 3D widget, which defines a spherical volume of interest in the data modeling space. Data contained in such a volume can be visualized with various modalities, according to suitable filters associated with MagicSphere. One such filter, the MultiRes filter, has been designed for visualizing multiresolution datasets.
A more sophisticated approach was designed for the TAn v.2 system, based on the interactive extraction of variable resolution meshes according to the spatial placement of a focus probe.
Most of these researches are carried out in cooperation with Leila De Floriani and
the University of Genova.
A system to represent and visualize scalar volume data at multiple resolution is presented. The system is built on a multiresolution model based on tetrahedral meshes with scattered vertices that can be obtained from any initial dataset. The model is built off-line through data simplification techniques, and stored in a compact data structure that supports fast on-line access. The system supports interactive visualization of a representation at an arbitrary level of resolution through isosurface and projective methods. The user can interactively adapt the quality of visualization to requirements of a specific application task, and to the performance of a specific hardware platform. Representations at different resolutions can be used together to enhance further interaction and performance through progressive and multiresolution rendering.
A scattered volumetric dataset is regarded as a sampled version of a scalar field defined over a three-dimensional domain, whose graph is a hypersurface embedded in a four-dimensional space. We propose a multiresolution model for the representation and visualization of such dataset, based on a decomposition of the three-dimensional domain into tetrahedra. Multiresolution is achieved through a sequence of tetrahedralizations that approximate the scalar field at increasing precision. The construction of the model is based on an adaptive incremental approach driven by the local coherence of the scalar field.
The proposed model allows an efficient extraction of compact isosurfaces with adaptive resolution levels as well as the development of progressive and multiresolution rendering approaches. Experimental evaluations of the proposed approach on different scattered datasets are reported.
How to render very complex datasets, and yet maintain interactive response times, is a hot topic in computer graphics. The MagicSphere idea originated as a solution to this problem, but its potential goes much further than this original scope. In fact, it has been designed as a very generical 3D widget: it defines a spherical volume of interest in the dataset modeling space. Then, several filters can be associated with the MagicSphere, which apply different visualization modalities to the data contained in the volume of interest. The visualization of multi-resolution datasets is selected here as a case study and an ad hoc filter has been designed, the MultiRes filter. Some results of a prototipal implementation are presented and discussed.