ROS catkin Wrapper around DSO: Direct Sparse Odometry

For more information see https://vision.in.tum.de/dso

This is meant as simple, minimal example of how to integrate DSO from a different project, and run it on real-time input data. It does not provide a full ROS interface (no reconfigure / pointcloud output / pose output). To access computed information in real-time, I recommend to implement your own Output3DWrapper; see the DSO code.

Related Papers

• Direct Sparse Odometry, J. Engel, V. Koltun, D. Cremers, In arXiv:1607.02565, 2016

• A Photometrically Calibrated Benchmark For Monocular Visual Odometry, J. Engel, V. Usenko, D. Cremers, In arXiv:1607.02555, 2016

1. Installation

1. run

    catkin_make
   

2. Usage

change the project folder name into dso_ros.

modify the bagfilename from src/main.cpp. you can modify this to roslaunch later.

	rosrun dso_ros dso_ros

3. Accessing Data.

see the DSO Readme. As of now, there is no default ROS-based Output3DWrapper - you will have to write your own.

4 License

This ROS wrapper around DSO is licensed under the GNU General Public License Version 3 (GPLv3). For commercial purposes, we also offer a professional version, see http://vision.in.tum.de/dso for details.