Skip to content

Instantly share code, notes, and snippets.

@amycheng

amycheng/analysis of BOX.md

Last active December 23, 2015 22:19
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save amycheng/6702892 to your computer and use it in GitHub Desktop.
Save amycheng/6702892 to your computer and use it in GitHub Desktop.
Download ZIP
a guess at how the BOX performance works
Raw
analysis of BOX.md

A quick and dirty technology dissection of BOX

The installation/performance in question: http://www.botndolly.com/box .

Bot and Dolly (the firm behind this performance/installation) will will apparently release a post-mortem at some point but here's a quick dissection of the tech.

A big part of this performance is video/projection mapping (mapping animations and graphics to physical objects that aren't necessarily a screen).

MadMapper (http://www.madmapper.com/) seems to be the app of choice to getting started with projection mapping really quickly is (it's expensive tho).

Other tools of choice for projection mapping (for those that can't afford pricey software) are: OpenFrameWorks and Cinder. Both are kinda like wrappers for C++ and OpenGl.

Step one is to map the geometry that you're projecting onto. This will entail either building a 3D computer model by hand, or using contour detection using OpenCV (an open source computer vision library).

Step two is to map animations onto the geometry.

Step three, using a high power projector and do your thing.

A precursory glance at the tech sheet (http://www.botndolly.com/_/uploads/robot_tearsheet/1_121108_070546.pdf) for the robotics system they are using, suggests that they are using Maya 3D to do everything: modeling, animation, timing. Maya 3D is a 3D animation workhorse.

Linkssss:

http://www.openframeworks.cc/

http://libcinder.org/

http://www.autodesk.com/products/autodesk-maya/overview

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment