Skip to content

Latest commit

 

History

History
64 lines (45 loc) · 2.04 KB

README.md

File metadata and controls

64 lines (45 loc) · 2.04 KB

Junkyard Jumbotron

The Junkyard Jumbotron lets you take a bunch of random displays and instantly stitch them together into a large, virtual display, simply by taking a photograph of them. It works with laptops, smartphones, tablets --- anything that runs a web browser. It also highlights a new way of connecting a large number of heterogenous devices to each other in the field, on an ad-hoc basis.

Here are instructions for installing the Junkyard Jumbotron. Most of these packages are available via apt-get on Linux, or in MacPorts or Homebrew of OS X (deployments are intended to be on Ubuntu or Debian boxes, and currently most development happens on OS X).

  1. Git

You'll need a git client to get the source code. Git is a distributed source code management system. You can get a client from the git homepage.

  1. Node & NPM

Junkyard Jumbotron runs on top of node.js (using it as a web application server). You need to install Node.js from http://nodejs.org/ NPM is a simple package management system for node packages. Follow their installation instructions.

  1. Python & Python Imaging

The Junkyard Jumbotron's image processing is all done in Python, utilizing a wrapper around the ARToolkitPlus augmented reality tracking library. You need to install Python 2.6 or 2.7 from their downloads page.

To support the image processing you also need the Python Imaging Library.

You also need to compile the ARToolkit, which is included in the repo:

> make python-extension
  1. GraphicsMagick

To support the image manipulation you also need the GraphicsMagick library.

  1. Node Packages

The Junkyard Jumbotron server relies on a number of node.js packages. You can install them all by running this command:

> make node-packages
  1. Run It!

To run it just do:

> node jjserver.js