Here are the installation instructions to set up the object_recognition package in ROS. We can install it using prebuilt binaries and source code. The easiest way to install is via binaries.
Here is the command to install ORK packages in ROS:
$ sudo apt-get install ros-kinetic-object-recognition-*
If you want to install these packages in ROS Indigo, replace kinetic with indigo.
This command will install the following ROS packages:
- object-recognition-core: This package contains tools to launch several recognition pipelines, train objects, and store models.
- object-recognition-linemod: This is an object recognition pipeline that uses linemod from OpenCV. The linemod pipeline is best for rigid body detection.
- object-recognition-tabletop: This is a pipeline use for pick-and-place operations from a flat surface
- object-recognition-tod: Textured Object Recognition is another pipeline for textured objects that uses features for detection.
- object-recognition-reconstruction: This is a basic 3D reconstruction of an object from aligned Kinect data.
- object-recognition-renderer: This is code that generates random views of an object.
- object-recognition-msgs: This package contains the ROS message and the actionlib definition used in object_recognition_core.
- object-recognition-capture: Capture is a set of tools to capture objects in 3D and perform odometry.
- object-recognition-transparent-objects: This is a technique to recognize and estimate poses of transparent objects.
- object-recognition-ros-visualization: This package contains Rviz plugins to visualize ORK detection results.
Here are the commands to install the packages from source. This command is basically based on the rosinstall tool, which helps set up a list of packages in a single command. You can run this commands from the /home/<user> folder.
$ mkdir ws && cd ws
$ wstool init src https://raw.github.com/wg-perception/object_recognition_core/master/doc/source/ork.rosinstall.kinetic.plus
$ cd src && wstool update -j8
$ cd .. && rosdep install --from-paths src -i -y
$ catkin_make
$ source devel/setup.bash
You can find more about LINE-MODE from the following link: http://far.in.tum.de/Main/StefanHinterstoisser This is the GitHub repository of the object-recognition packages: https://github.com/wg-perception Here are the possible issues you may have while working with this package: https://github.com/wg-perception/object_recognition_ros/issues https://github.com/wg-perception/linemod/issues https://github.com/wg-perception/object_recognition_core/issues