our work (fwd)

STUART JOSHUA MICHAEL (stuartj@rintintin.Colorado.EDU)
Tue, 7 Apr 1998 12:08:02 -0600 (MDT)

Just wanted to include this message I previously sent to Andrea Synyder
privately in case anyone else was curious about the kind of work we're
doing here in Colorado.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
josh stuart
professional research assistant
dr. elizabeth bradley's lab
department of computer science
university of colorado
stuartj@cs.colorado.edu
(303) 492-8425
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

---------- Forwarded message ----------
Date: Mon, 6 Apr 1998 12:43:08 -0600 (MDT)
From: STUART JOSHUA MICHAEL <stuartj@rintintin.Colorado.EDU>
To: "Snyder, Andrea" <ASnyder@MAIL.Kennedy-Center.org>
Subject: our work

Hi Andrea!

Thanks for expressing interest in our work! I'm sorry I didn't respond to
your email sooner; I was taking a mini ski-vacation last week.

For the past year, we have been developing methods of shuffling dance using
chaotic sets of equations. Our hope was this temporal rearrangement
of body positions would give choreographers ideas for new variations
and combinations derived from pre-choreographed routines. However, one result
of the shuffling is that impossible (or stylistically illegal) moves are
introduced into the newly generated dance and we wanted to find some way
to correct these bad movements. For example, after the shuffling is complete,
it is possible the dancer is kicking her left leg in one frame and in the
next frame is lying flat on her back. We want the computer to introduce
new key frames between two incompatible ones that make the sequence look
plausible. For instance, if we gave the computer the previous example as
input, we would want the computer to drop the left leg, make the dancer squat
to sit on the ground, and finally lean back into a lying position. You can
find a few of our papers on Dr. Bradley's homepage:

http://www.cs.colorado.edu/~lizb/Home.html

look under "Recent Publications".

We have been using LifeForms primarily to view the sequences the shuffler
and the interpolator come up with. What I do is I take an animation, convert
it to a script file (an ASCII text file that my programs can read), shuffle
this script file to produce a another script file, and then import this new
script file back into lifeforms to view it. It's rather tedious to be honest.
We are working on our own web-based figure animator at the moment so that
people can shuffle their sequences on the web.

I hope that answers some of your questions. Let me know if you have any
more. Also, since you're currently the only one who has answered my plea
for LifeForms' animations, please let me know if you know of anyone that can
help.

Thanks!

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
josh stuart
professional research assistant
dr. elizabeth bradley's lab
department of computer science
university of colorado
stuartj@cs.colorado.edu
(303) 492-8425
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++