Interactive Objects Learning

• YARP (with LUA bindings compiled)
• iCub
• icub-contrib-common
• OpenCV (3.3.0 or higher with tracking features enabled)
  1. Download OpenCV: git clone https://github.com/opencv/opencv.git.
  2. Checkout the correct branch/tag: e.g. git checkout 3.3.0.
  3. Download the external modules: git clone https://github.com/opencv/opencv_contrib.git.
  4. Checkout the same branch/tag: e.g. git checkout 3.3.0.
  5. Configure OpenCV by filling in OPENCV_EXTRA_MODULES_PATH with the path to opencv_contrib/modules and then toggling on all possible modules.
  6. Compile OpenCV.
• LUA
• rfsmTools
• segmentation
• Hierarchical Image Representation
• stereo-vision
• speech
• Multiple Instance Boosting (optional, required to compile milClassifier)

Remember to export the environment variable LUA_PATH with paths to lua scripts located in iol directory and put them also in the PATH.
Example:

• export LUA_PATH=";;;$ICUBcontrib_DIR/share/ICUBcontrib/contexts/iol/lua/?.lua"
• export PATH=$PATH:$ICUBcontrib_DIR/share/ICUBcontrib/contexts/iol/lua

Online documentation is available here: http://robotology.github.com/iol.

A video showing the recognition and interaction capabilities achieved by means of IOL components can be seen here.

Material included here is Copyright of iCub Facility - Istituto Italiano di Tecnologia and is released under the terms of the GPL v2.0 or later. See the file LICENSE for details.