Lorentz Center - VIEW: Visual Interactive Effective Worlds from 25 Jun 2007 through 27 Jun 2007
  Current Workshop  |   Overview   Back  |   Home   |   Search   |     

    VIEW: Visual Interactive Effective Worlds
    from 25 Jun 2007 through 27 Jun 2007

When To Walk Away: Questions To Ask In Infovis Projects
When To Walk Away: Questions To Ask In Infovis Projects
Tamara Munzner, 
University of British Columbia, Vancouver, Canada
I will discuss several questions that I recommend asking before an
infovis project starts, and re-asking at intervals throughout the
lifetime of the project. In this talk, I will not focus on technique
questions, such whether a technique is novel or whether a chosen
visual representation actually communicates the desired structure.
Instead, these questions are concerned with process issues. What
flavor of collaborators do I have - real users, or fellow tool
builders? Is the problem solveable: Is there a real need for my new
approach/tool? Am I addressing a real task? Does real data exist and
can I have it? I will give examples of several past projects, with
outcomes along the range from successful to not so successful, where
these questions mattered. In many cases, it did not occur to me to ask
these questions when starting the projects, even if they seem obvious
in retrospect.



Flow Visualization

Daniel Weiskopf

Simon Fraser University, Burnaby, Canada

The visualization of flow (e.g. fluid flow from the aerospace
and automotive industries) is a well known topic of scientific
visualization. In this talk, I give a brief overview of flow
visualization approaches in general and discuss some recent advances
in texture-based flow visualization in particular, where I focus on
the aspects of efficiency and image quality. This talk also covers
open questions and challenges, which include the visualization of
3D data, time-dependent flow, and large data sets.



Interactive Visualization of Diffusion Image Data and its Models
Gordon Kindlmann
Harvard Medical School, Boston, Massachusetts, USA

Diffusion Tensor MRI has become a popular way of non-invasively assessing micro-structural orientation and organization in biological tissue, especially the central nervous system.  Diffusion tensor data is in fact the result of a model-fitting post-process run on the original diffusion-weighted image (DWI) data acquired by MRI.  The single tensor fit of the DWI data has enabled many scientific and medical applications of DWI (e.g., tractographic connectivity studies, region-of-interest anisotropy measurements), but a single tensor is not the only way to model DWI data.  Some recent work, for example, has explored fitting two tensors per voxel.  With more complicated models, however, come more complicated algorithms for doing the model fitting.  Visualization can play a role in understanding the behavior of DWI modeling algorithms, so that the relationship between known anatomy, underlying DWI data, and estimated models can be explored in a quantitative but intuitive way.  This talk will describe new software for interactively visualizing DWI data and its models with the goal of better understanding the properties and potential of DWI data.




3D Visualization of Vasculature

Bernhard Preim
Otto-von-Guericke-Universität Magdeburg
Fakultät für Informatik/ Institut für Simulation und Graphik
Magdeburg, Germany

"A large variety of techniques has been developed to visualize vascular structures.
These techniques differ in the necessary preprocessing effort, in the computational effort to create the visualizations, in the accuracy with respect to the underlying image data and in the visual quality of the result. In this overview, 3D visualization methods are compared and their applicability for diagnosis, therapy planning and educational  purposes is discussed. In particular, model-based approaches, which rely on model assumptions to create “idealized" easy-to-interpret visualizations and model-free  approaches, which represent the data more faithfully, are distinguished. Furthermore, we discuss interaction techniques to explore vascular structures and illustrative techniques which map additional information on a vascular tree, such as distance to a tumor.
Despite the diversity and number of existing methods, there is still a demand for future research which is also discussed."



Virtual Reality & Presence

Frederic Vexo

Ecole Polytechnique Federale de Lausanne, Switzerland

The challenges of creating visual interactive contents seems less on the synthesis of

images  than on the quality of the relation between the user and digital contents. An

estimator of the quality of such relation is the level of presence which could be seen as a

evaluation of the feeling to be in a virtual worlds. This presentation aims to explain the

links between presence and virtual worlds and discuss some  strategies, methods and

techniques reinforcing the personal involvement of the users. Several research

examples will presented as support for the discussion.



Some Aspects of Optimal Sampling
Torsten Möller
Simon Fraser University, Burnaby, Canada
In this talk I will briefly derive the main idea behind optimal sampling distributions. This
leads to the body-centered cubic (BCC) lattice in 3D. I will then summarize our research
on the BCC lattice and highlight its advantages. We found it to have better numerical 
accuracy, as well as better perceptual accuracy. Further, it performs twice as fast as the
ubiquitous cartesian lattice for comparable ray-casting implementations.
I will conclude my talk with some startling observations about optimal sampling in high-
dimensional spaces.
Experimental Research in Visualization?! Creating a Common Research Focus
Jean-Bernard Martens
Eindhoven University of Technology, Department of Industrial Design, The Netherlands
There is a growing awareness that research in visualization could profit
from more experimental (also called empirical) research. This talk
addresses three related issues:
1. Why should researchers in visualization bother about experimental
research in the first place? It is argued that experimental research is
not a goal in itself but follows naturally from a wish for a more common
research focus, in which there is an agreement on important tasks and
ways of measuring performance in these tasks.
2. What are the main pitfalls when doing experimental research?
Especially the threats posed to reaching valid conclusions (also known
as the validity of a test) are discussed in more details.
3. Does one really need statistics and what is the (lack of) meaning of
a statistically significant effect? It is explained how statistical
modeling and visualization can support a more intuitive interpretation
of experimentally obtained measurements.