Skip to Main content Skip to Navigation

Hiérarchisation et visualisation multirésolution de résultats issus de codes de simulation

Fabien Vivodtzev 1
1 EVASION - Virtual environments for animation and image synthesis of natural objects
GRAVIR - IMAG - Laboratoire d'informatique GRAphique, VIsion et Robotique de Grenoble, Inria Grenoble - Rhône-Alpes, CNRS - Centre National de la Recherche Scientifique : FR71
Abstract : Numerical simulations generate large amounts of data that is far greater than the available resources on a workstation and it is unlikely that future developments of resource technology will be able to keep up with the growing demand. The finite element meshes supporting these simulations are normally composed of several millions of volumetric cells having embedded sub-structures of varying dimensions (surfaces and linear features) such as thin material-boundary layers. Postprocessing this large data in order to reduce the required amount of data needed to represent it allows a user to visualize and manipulate the data interactively on a single workstation. However, existing tools in scientific visualization do not allow or only partially reach the interactive exploration goals for this type of data (i.e., large number of cells, sub-structures, thin layers . . .).

The dissertation presents two main steps for interactively rendering large data ; a postprocessing and visualization step. The postprocessing step uses an approach having hierarchical organization of data in order to construct a multiresolution representation allowing subsequent interactive visualization. This step is based on a mesh simplification algorithm that uses iterative edge collapses while preserving both mesh topology and the topology of all embedded sub-structures. The robust topological criteria introduced in this work are derived from theoretical notions in algebraic topology. The visualization step uses the multiresolution representation produced in the first step to speed up rendering time of the data. In a progressive, invertible and local way, the method adapts dynamically to the required resolution, specified by the user, and the hardware resources available.

This dissertation illustrates the application of these techniques of hierarchical organization and visualization on data from various fields, particularly on data obtained from electromagnetic simulations by the CEA/CESTA.
Complete list of metadata

Cited literature [40 references]  Display  Hide  Download
Contributor : Fabien Vivodtzev <>
Submitted on : Saturday, December 17, 2005 - 3:05:17 PM
Last modification on : Monday, December 28, 2020 - 3:44:02 PM
Long-term archiving on: : Saturday, April 3, 2010 - 7:33:18 PM


  • HAL Id : tel-00011230, version 1



Fabien Vivodtzev. Hiérarchisation et visualisation multirésolution de résultats issus de codes de simulation. Interface homme-machine [cs.HC]. Université Joseph-Fourier - Grenoble I, 2005. Français. ⟨tel-00011230⟩



Record views


Files downloads