Project details can be found at http://rbart.github.com/guide-dog/

We have successfully run Guide-Dog on both Ubuntu and Fedora using the following directions:

• Installing libraries for Guide-Dog on Ubuntu Linux version 12.04.1 LTS (recommended OS)

  • Open a Terminal and execute (these instructions taken from http://pointclouds.org/downloads/linux.html ):

    sudo add-apt-repository ppa:v-launchpad-jochen-sprickerhof-de/pcl
    sudo apt-get update
    sudo apt-get install libpcl-all
    

  • Download and extract a copy of the OpenCV source code from http://opencv.willowgarage.com/wiki/

  • Change directory into the OpenCV-2.4.2 directory.

  • Execute mkdir build

  • Change directory to build.

  • Execute cmake ... You'll likely need to install other packages required by OpenCV.

  • After cmake has configured successfully, execute sudo make install.

  • Install OpenAL and AL Utility Toolkit from the respository with sudo apt-get install libopenal-dev libalut-dev

• Installing libaries for Guide-Dog on Fedora Linux 17

  • Install PCL from the repository with the following command sudo yum install pcl pcl-devel pcl-tools pcl-doc
  • Install OpenNI and the Primesense driver from the repository with sudo yum install openni-devel openni-primesense
  • Install OpenCV from the repository with the following command sudo yum install opencv-devel
  • Install OpenAL and AL Utility Toolkit with the following command sudo yum install openal-soft-devel freealut-devel

• Now let's get the source code to compile Guide-Dog

  • Download and extract a copy of the PCL source code from http://www.pointclouds.org/assets/files/1.6.0/PCL-1.6.0-Source.tar.bz2
  • Download Guide-Dog to the same parent directory as the PCL source code by running git clone https://github.com/rbart/guide-dog.git
  • From the parent directory that contains the guide-dog directory and PCL-1.6.0-Source directory execute cp -R guide-dog/sonic-dog/ PCL-1.6.0-Source/; cp -R guide-dog/combined/visualization/ PCL-1.6.0-Source/
  • Change directory to the PCL-1.6.0-Source directory.
  • Execute mkdir build.
  • Change directory to build.
  • Execute cmake .. and make sure that the visualization package will be built. You'll likely need to install other packages required by PCL.
  • Execute make openni_viewer.
  • Copy the .wav files under guide-dog/sonic-dog to 'build'.
  • Make sure your camera is plugged in to a USB port. We recommend the Asus Xtion PRO LIVE. If possible, use a USB 2.0 port.
  • Execute ./bin/openni_viewer to run Guide-Dog.