SciVisFall2007/Schedule

8/21: Introduction to visualization

Lecturer: Claudio

Topics: Scientific Visualization

Animations: explosion_640x480-5.mov, fig7.mov, fig8.mov, fig9.mov, SevereTstorm.mov

Further reading:

Visualizing Spatial and Temporal Variability in Coastal Observatories, W. Herrera-Jimenez, W. Correa, C. Silva, and A. Baptista, IEEE Visualization 2003.

8/23: The visualization pipeline

Lecturer: Claudio

Topics: Procedural vs. Dataflow programming; Using Dataflow for the Vis Pipeline; Dataflow programming with VTK; Dataflow programming with VisTrails; python.

Notes: lec02-notes.pdf

Slides: lec02.pdf

VisTrails: During this class, we built a pipeline equivalent to the cone.tcl (see class slides). Here is the vistrails file: cone.vt

Further reading:

Provenance for Visualizations: Reproducibility and Beyond, C. Silva, J. Freire, and S. Callahan, IEEE Computing in Science and Engineering, to appear.

8/28: Modeling Data for Visualization

Lecturer: Carlos

Topics: Discrete vs continous data; Sampling and interpolation; Point vs triangulated data; Meshing data types; Regular vs irregular data; Tabular data; Vector and tensor fields

Notes: TBA

Slides: .ppt file

Further reading:

There is no required reading for this lecture. The notes will be available shortly. The following papers are there for people that are looking to get more advanced material that will not be covered in class.

Interpolation

Geodesic-loxodromes... This is the fancy interpolation for diffusion tensors I mentioned in class.

Bernstein polynomials These are the polynomials used for cubic Bezier curves that I mentioned in class.

Separability

Extensions of the Zwart-Powell Box spline... This is a recent paper that shows a class of trivariate reconstruction techniques that are not separable.

Tensors

Visualization and Analysis of Diffusion Tensor Fields Gordon Kindlmann's PhD. thesis, with everything you ever wanted to know about DTI. Section 2.1 has a good primer in tensor algebra.

8/30: Modeling Data for Visualization

Lecturer: Carlos

Topics: Geometry Processing: Reconstruction and meshing; Simplification; Smoothing; Other Filtering algorithms

Notes: .pdf file. If you want to print these, you might want to wait for a week or two, until I finish polishing them.

Slides: .ppt file These slides include simplificatin algorithms, which I'll talk about next week.

9/4: Elementary Plotting Techniques

Lecturer: Steve

Topics: Principles of Graph Construction

Notes: PlottingNotes.pdf

Slides: Plotting1.pdf

Vistrails: See next lecture.

Further Reading: There is no required reading for this lecture. For those interested in more depth, the following books are very useful:

The Elements of Graphing Data. William S. Cleveland, Hobart Press, 2nd Edition, 1994.
Visualizing Data. William S. Cleveland, Hobart Press, 1993.
The Visual Display of Quantitative Information. Edward R. Tufte, Graphics Press, 2001.
Visual Explanations: Images and Quantities, Evidence and Narrative. Edward R. Tufte, Graphics Press, 2997.

9/6: Elementary Plotting Techniques

Lecturer: Steve

Topics: Simple Plotting Methods: Dot Plots, Connected Symbol Plots, Scatter Plots, Histograms, Others. Advanced Plotting Methods: Multimodal, Higher Dimensional, Correlation, Uncertainty and Variation.

Notes: PlottingNotes.pdf

Slides: Plotting2.pdf

Vistrails: PlottingVistrails.zip - Unzip this file in the examples directory of your VisTrails installation and it will add the vistrails along with their data sets (in the data directory). If you don't have permission to write to this directory (CADE users), then unzip the file where you want. Just be aware that in this case the paths for the data files may not be correct for most vistrails and will need to be fixed before they will execute properly.

Further Reading: There is no required reading for this lecture. Some articles of interest:

9/11: Color and Human Perception

Lecturer: Claudio

Topics: Human vision system; Optical illusions

Notes: TBA

Slides: human-vision.pdf

Links:

http://en.wikipedia.org/wiki/Eye

http://www.grand-illusions.com/gregory2.htm (also, see the related book: [1])

http://en.wikipedia.org/wiki/Purkinje_effect

http://www.handprint.com/HP/WCL/color2.html

9/13: Color and Human Perception

Lecturer: Claudio

Topics: Color Science; Color spaces; Color Blindness; Color maps; Tone mapping

Notes: TBA

Slides: colorvision.pdf

Links:

Further reading:

How Not to Lie with Visualization

http://en.wikipedia.org/wiki/Opponent_process

http://en.wikipedia.org/wiki/Color_models

http://en.wikipedia.org/wiki/Absolute_color_space

http://en.wikipedia.org/wiki/Additive_color

http://en.wikipedia.org/wiki/Subtractive_color

http://en.wikipedia.org/wiki/RGB_color_model

http://en.wikipedia.org/wiki/SRGB_color_space

http://en.wikipedia.org/wiki/CIE_XYZ_color_space

9/18 (a): Color and Human Perception

Lecturer: Claudio

Same material as previous lecture.

9/18 (b): 2D Visualization Techniques

Lecturer: Carlos

Topics: 2-D contours, marching quads, marching tris; Color mapping; height fields; NPR

Slides: pdf file

Notes: pdf file

Vistrails: zip file with ozone.vt and data asymptotic decider in 2d heightfields

Note: These vistrails use relative file paths so you don't need to change each of them individually to match your directory structure. Simply unzip the file to whichever location is more convenient. Then, inside VisTrails, open the VisTrails shell, type:

import os
os.chdir("c:/directory/where/you/unzipped/it")

This will change the directory so you should be able to just run the pipelines.

9/20: Math refresher

Lecturer: Carlos

Topics: Basic linear algebra; vectors; basic differential geometry (space curves, tangents, normals, surfaces); basic vector calculus (gradient, divergence, curl, gauss' theorem, green's theorem)

9/25: 2D Visualization Techniques

Lecturer: Carlos

Topics: 2-D vector fields, div, grad, curl in 2D; Steady vs Unsteady flows; Glyphs; 2-D streamlines, streaklines, pathlines

Slides: pdf file

Notes: coming soon

Vistrails: vistrail with steady vector field vis and data vistrail with unsteady vector field vis and data Note: Because VTK does not support time-varying datasets directly, we had to create a reasonably ugly hack to simulate unsteady fields. This means the datasets are quite big (80MB in total).