Particle based tensor field visualization.

Gregoire Piquet and Thomas Viard and Guillaume Caumon. ( 2010 )
in: Proc. 30th Gocad Meeting, Nancy

Abstract

Many physical quantities relevant in geomodeling consist of second order tensors (e.g. permeability, strain). However graphic tools are often not specifically designed for their visualization. New tensor visualization methods would therefore greatly facilitate the interpretation and characterization of geological objects. Even if the visualization of 3-D scalar fields has been extensively studied and is routinely used in geosciences, 3-D tensor fields are still commonly visualized in a suboptimal way by looking at only some invariants (eigen values and vectors). Several alternative approaches have been proposed in the literature, but are not yet standard in geosciences. Meanwhile, programmable graphics cards open new avenues for addressing these problems and efficiently generate meaningful and more understandable representations of 3-D tensor fields. The aim of this project is to propose a new method that allows tensor field visualization. The method is based on a particle engine algorithm which represents the tensor attributes as specific particle behaviors. For instance, the particle velocity depends on the amplitude of an eigen value of the tensor, and the particle direction follows the corresponding eigen vector. The particles are continuously generated at each frame. This helps the user identifying pseudo-flow paths and digitations, and to quickly track pseudo-streamlines from entry points selected randomly on the surface. An option is to map the color of particles to an underlying scalar field, allowing for adding one more dimension to the display. Moreover, it is hoped that this new functionality would help the Gocad users in their interpretation of tensor fields such as permeability, deformation and stress.

Download / Links

BibTeX Reference

@inproceedings{PiquetGM2010,
 abstract = { Many physical quantities relevant in geomodeling consist of second order tensors (e.g. permeability, strain). However graphic tools are often not specifically designed for their visualization. New tensor visualization methods would therefore greatly facilitate the interpretation and characterization of geological objects.
Even if the visualization of 3-D scalar fields has been extensively studied and is routinely used in geosciences, 3-D tensor fields are still commonly visualized in a suboptimal way by looking at only some invariants (eigen values and vectors). Several alternative approaches have been proposed in the literature, but are not yet standard in geosciences. Meanwhile, programmable graphics cards open new avenues for addressing these problems and efficiently generate meaningful and more understandable representations of 3-D tensor fields.
The aim of this project is to propose a new method that allows tensor field visualization. The method is based on a particle engine algorithm which represents the tensor attributes as specific particle behaviors. For instance, the particle velocity depends on the amplitude of an eigen value of the tensor, and the particle direction follows the corresponding eigen vector. The particles are continuously generated at each frame. This helps the user identifying pseudo-flow paths and digitations, and to quickly track pseudo-streamlines from entry points selected randomly on the surface.
An option is to map the color of particles to an underlying scalar field, allowing for adding one more dimension to the display. Moreover, it is hoped that this new functionality would help the Gocad users in their interpretation of tensor fields such as permeability, deformation and stress. },
 author = { Piquet, Gregoire AND Viard, Thomas AND Caumon, Guillaume },
 booktitle = { Proc. 30th Gocad Meeting, Nancy },
 title = { Particle based tensor field visualization. },
 year = { 2010 }
}