Giving Human Form to Audio Visualizers, Version 1

I'll upload a video of this audio visualization when I have some time. In the meantime, try the code out yourself. I'm still tweaking it to figure out what might look best.

Some requirements to run this code:
1. Download OpenCV for Processing from here: https://github.com/atduskgreg/opencv-processing/releases
2. And download and install from here: http://opencv.org/
2.b. Probably need v2.4.5 of opencv from opencv.org, but I was able to get it to work with v2.4.9

I'm also using processing 2.1.2.

And finally, copy this gist into processing: