Skip to content
Titus Cieslewski edited this page Jan 7, 2016 · 5 revisions

Run SVO on a Dataset

Download this example dataset: rpg.ifi.uzh.ch/datasets/airground_rig_s3_2013-03-18_21-38-48.bag

Open a new console and start SVO with the prepared launchfile:

roslaunch svo_ros test_rig3.launch

Start RVIZ (Robot Visualizer) in a new console

rosrun rviz rviz -d <PATH TO rpg_svo>/svo_ros/rviz_config.rviz

Now you are ready to start the rosbag. Open a new console and change to the directory where you have downloaded the example dataset. Then type:

rosbag play airground_rig_s3_2013-03-18_21-38-48.bag

Now you should see the video with tracked features (green) and in RViz how the camera moves. If you want to see the number of tracked features, fps and tracking quality, run the GUI.

Run SVO on a live camera stream

Assuming that you have a ros node that publishes images on the topic camera/image.

  1. Calibrate your camera (see Camera Calibration) and save a new calibration file in svo_ros/params/my_camera.yaml.
  2. Adapt the launchfile in svo_ros/launch/live.launch: Change the topic and the link to the new calibration file.
  3. Start SVO using roslaunch svo_ros live.launch

SVO GUI

Type rosrun rqt_svo rqt_svo to run the SVO widget that displays the number of tracked features, the frame rate and provides some interface buttons. If you do not see any reaction from GUI, you can type svo in namespace column(which is default name of this package).

If the widget is not found, try this:

rm ~/.config/ros.org/rqt_gui.ini
rosrun rqt_svo rqt_svo

Keyboard Shortcuts

Make sure to active the console window when pressing the keys.

  • s Start/Restart
  • q Quit
  • r Reset