This is some additional material for mac and linux users.
For the windows peoplem I would ssuggest either install linux or go for visual studio with the primesence develpoment kit.
Either way you need to be able to access to the raw data coming from the kinect (RGB and depth maps).
WARNING THE SECTION BELOW HAS CHANGED
- you need to have libfreenect working on your machine (cf. http://openkinect.org/wiki/Getting_Started)
- for ubuntu it runs in 4 commands:
- sudo add-apt-repository ppa:arne-alamut/freenect
- sudo apt-get update
- sudo apt-get install freenect
- sudo adduser YOURNAME video
- To start the demo applications: freenect-glview
- for mac (assuming you installed port and xcode development kit):
- sudo port install git-core
- sudo port install libtool
- sudo port install libusb-devel
- git clone https://github.com/OpenKinect/libfreenect.git
- cd libfreenect/
- mkdir build
- cd build
- cmake .. && make && sudo make install
- To start the demo run freenect-glview
xLearn and Lua:
All the packages we will be working on during class have been included in the xLean git repo.
For those of you who will be using Lua with
libfreenect you will have to compile everything from the sources
in order to be able to edit the code and make it run. The process
should be much simpler than last week. (sorry for the confusion)
From now on we will consider that you have clone the git repository in the xlearn folder
if you have followed all those instructions you should be able to run this in the qlua:
Lua 5.1 Copyright (C) 1994-2008 Lua.org, PUC-Rio
> require 'mincut'
In addition there is a demo with a GUI to play with
the thresholding. To run it you need to go in the demos folder (located
in the xlearn repo that you have cloned).
For this homework you need to play with the segmentation and understand what is hapenning.
If you are using Lua, all the code is here: xlearn/torch/contrib/mincut/
whenever you modify the code you need to rebuild everything, here is the command for rebuild:
cd xlearn (go to the root of the repo)
sudo make update
If you have any questions, please use the mailing list email@example.com