These pages no longer maintained
Please visit http://wwwx.cs..unc.edu/~jwendt/wordpress for current webpage. Also, please update all links.
Graduate Years
2003-present
Graduate school has been (and continues to be) quite an adventure! After
the long drive from Provo, UT to Chapel Hill, NC, I began researching with
Bill Baxter and Ming Lin. Due to the mixture of research work and study
in graduate school, I will be following a slightly different organization
here than I did on my Undergraduate Page.
SUMMER 2003
Classes
No classes taken during summer term
Research
Starting June 2003, I was a member of UNC's
GAMMA Group headed by
Ming C. Lin and
Dinesh Manocha. Ming was my
official advisor, and I began by working with
Bill Baxter. Bill was working
on extending dAb into a more
physically accurate model. I was assigned the responsibility of
improving the rendering system. Graphics cards had recently become
more programmable, so my first task was to learn
nVIDIA's Cg.
I ordered the The Cg Tutorial: The Definitive Guide to
Programmable Real-Time Graphics by Fernando and Kilgard. I became
a fairly good GPU programmer. I also looked into the Kubelka-Munk
(KM) reflectance model
(see
This reference for a brief intro. I would actually suggest just reading
the original paper as referenced on that page). Our system ended up using
an implementation of KM reflectance on the GPU. I spent the summer learning
about these things, preparing for a paper deadline.
FALL 2003
Classes
COMP 202 - Algorithm Analysis
Taught by James Anderson.
Text: Introduction to Algorithms, 2nd ed. by Cormen, Leiserson, Rivest and Stein
This class was pure paper and pencil. Jim required us to be very clear and
complete in our proofs. We covered the following chapters in depth:
7 Quicksort, 18 B-Trees, 19 Binomial Heaps, 21 Data Structures for Disjoint
Sets, 24 (Review) Dijkstra's and Bellman-Ford Algorithms, 25 All-Pairs
Shortest Paths, 26 Maximum Flow, 34 NP-Complete and Related Problem Classes,
35 Approximation Algorithms; and the following in less depth: 29 Linear
Programming, 30 Polynomials and the FFT, 32 String Matching. Busy class.
COMP 281 - Computational Geometry
Taught by Jack Snoeyink.
Text: Computational Geometry, 2nd ed. by deBerg, vanKreveld, Overmars and Schwartzkopf
Again, pure paper and pencil (with one small programming assignment). We
went through the text almost completely, although we didn't go from beginning
to end (i.e. we jumped around a lot). The areas that I have used the most
since completing the course are Voronoi Diagrams, Duals, Delauney
Triangulations, and Convex Hulls.
Research
Continued working with Bill Baxter and Ming Lin. We completed our work on
what we called IMPaSTo in time
to submit to Non-Photorealistic Animation
and Rendering, 2004. Our paper was accepted and can be found at IMPaSTo
link provided above. My contribution to this work was the rendering pipeline,
measurements of the paint, the video, webpage, and section 5 of the paper.
While I would follow up some of this work before the
photo-ready version of the paper was due, I was done with the "development"
side of this work by the end of October. BTW, my wife, Heather, painted the
Dame en Blau that graces the front page of the paper.
During November, I developed the
code page that can be found on my website. I've mostly stuck with FLTK
since then.
SPRING 2004
Classes
COMP 259 - Physically Based Modelling
Taught by Ming C. Lin.
Text: None.
Please follow the above link for my description and course projects.
MATH 192 - Scientific Computation 2
Taught by David Adalsteinsson.
Text: Applied Numerical Linear Algebra by James Demmel
This class was a split between pencil and paper and MATLAB scripts and small
bits of C with write-ups. Several of my write-ups were email only, so they
aren't listed. If you follow this link you will go to
the directory with the projects in it. Math192Code is code provided by David.
The others are small projects I coded up (generally with Kimberley Noonan as
partner) as assignments. These are provided AS IS, with write-ups sometimes,
images of convergence usually, and the code as it was when I turned it in. I
make no promises of usefulness to people trying to implement real world
things.
COMP 321 - Technical Communication in Computer Science
Taught by Fred Brooks and
Stephen Weiss.
Text: Teaching Tips by McKeachie
We discussed several different teaching goals, as well as each person giving
a presentation that was critiqued by the class. My
presentation was about the migration of the "Mormon" people from Nauvoo,
IL to the Salt Lake Valley in Utah.
Research
I finished the previous semester and began this one with a big hard push for
SIGGRAPH 2004 with Brandon Lloyd.
We completed and submitted a work entitled
CC Shadow Volumes which increased
running time of shadow volumes by clamping the shadow volumes to only the
area where they shadowed a piece of visible geometry. This reduced the
time required to write to the stencil buffer in some heavily shadowed scenes.
Although this was not accepted to SIGGRAPH 2004, it was published at
Eurographics Rendering Workshop. My contributions here were in parsing the
input file and turning it into the format Brandon wanted, revising and adding
images to the paper, creating the video and webpage. I understand the
algorithm, but did not help in coding it.
I spent the remainder of the semester on two things: physically-based
deformation (which ended up being my final project in COMP
259) and the birth
of my son.
SUMMER 2004
Classes
No classes taken.
Research
My research into deformation in computer graphics showed me that as incredible
as Free Form Deformation is, Finite Element Methods were the wave of the
future. I spent the summer reading various texts on Finite Element Methods
with an emphasis on deformation. Among the texts that I read part of were:
Non Linear Elastic Deformations by R. W. Ogden, Nonlinear Continuum
Mechanics for Finite Element Analysis by Bonet and Wood, Numerical
Solution of Partial Differential Equations by the Finite Element Method
by Claes Johnson, A Unified Approach to the Finite Element Method and
Error Analysis Procedures by John O. Dow, and The Finite Element
Method, Linear Static and Dynamic Finite Element Analysis by Thomas J. R.
Hughes. For the beginner, the last book is probably the easier to understand,
and it is the one that I studied the most. I coded several simple FEM solvers
for 1D Static, 2D (1D out) Static, 2D (2D out) Static and 2D (2D out) Dynamic
problems. I can provide code and/or executables to interested people if they
contact me. I freely give the
lecture slides which I
prepared for a class lecture on FEM.
FALL 2004
Classes
COMP 204 - Software Design and Implemenation
Taught by David Stotts.
Text: Design Patterns: Elements of Reusable Object-Oriented Software by
Gamma, Helm, Johnson and Vlissides (AKA Gang-of-Four (GoF) book)
This class was mostly pencil and paper. The link provided above is the
course webpage by David Stotts, and is likely being updated every other
year to reflect the current incantation of the class. I did do a small webpage
which I will link here. However, I fully expect that
no one will cheat off of this page by copying any of my discussion or other
work. If I find that any of this has been swiped, I will remove the page all
together.
COMP 238 - Advanced Image Synthesis
Taught by Anselmo Lastra
Text: None
My work for this class can be found here. We wrote
and analyzed a ray tracer, as well as learning Pixar's Renderman.
MATH 221 - Numerical ODE/PDE 1
Taught by Michael L. Minion
Text: None required, any number of books recommended.
This class was a mix of pencil and paper and MATLAB/C codes. I provide my
write-ups, code, etc. with the same provisions as above: no cheating, and that
these are quick learning codes, so they are not perfect, nor necessarily
fully correct. They are found here.
Research
SIGGRAPH 2004's paper by Mark
Carlson, Rigid Fluid, really made me wonder about fluid/deformable
interactions. I spent this semester investigating how that might be done.
The first step I decided on was that Finite Difference Methods were no longer
the best way to simulate fluids. I have a paper under review about this now
so I'll hold off on saying more until I know the result.
WINTER 2005
Classes
COMP 242 - Operating Systems
Taught by Prasun Dewan.
Text: Operating System Design, Vol. 1, The Xinu Approach by Comer and Fossum
This class was heavily a coding course. We implemented a XINU-based
Distributed (light) Operating System. We actually wrote a Linux process which
handled its own timing interruptions, queue of "applications", etc. and could
communicate with other XINU/Linux Processes on this and other machines. No
code is posted here because I'm certain Prasun doesn't want it posted.
MATH 222 - Numerical ODE/PDE 2
Taught by Michael L. Minion
Text: None required, any number of books recommended.
Again a mix of pencil, MATLAB and C. Here are some more
write-ups. We worked on both compressible and
incompressible fluids and the numerical issues of both. Note that the point
in these classes was to analyze how the PDE solvers converged, not to
implement beautiful systems.
Research
More of the "secret" stuff. That will arrive here as soon as I know the
results of the submission.
SUMMER 2005
Classes
None.
Research
As luck would have it, I switched research areas this summer. I am now
working with EVE - Effective
Virtual Environments, where we create virtual worlds in order to study
the effectiveness of various visual, audio, and tactile techniques. I am
advised by Fred Brooks and
Mary Whitton.
This summer I will read several texts. Among them: Multiple View
Geometry in Computer Vision by Hartley and Zisserman.
Back to Main Index