for those of you interested (looking at you Andrew Schreiber), I made some changes to my min_fn and used a constant environment. The results can be found
here.
The student who I took under my wing this past build season has already implemented 254's code to make a segmented path into curves, the results can be found
here. What's really cool is that apparently my program's output is suitable for 254's code to run on. I need to combine these two and see how long it takes to run through completely.
This is still a major work in progress, but it is getting somewhere.
The next step is to apply depth map readings into the party.
All this is in my dropbox of
computer vision stuff