Skip to content

PaoPaoRobot/docker_dvo

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

23 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Dense Visual Odometry (dvo)

These packages provide an implementation of the rigid body motion estimation of an RGB-D camera from consecutive images.

Usage

You can use my docker image

docker pull paopaorobot/dvo

You need mount your data to docker container, you can download TUM data from here

Before the next step, you need get the assoc.txt file by run:

python2 associate.py rgb.txt depth.txt >> assoc.txt

Then, you can run

docker run -i -t -p 5900:5900 -e RESOLUTION=[width]x[hight] -v [data_path]:[docker_data_path] paopaorobot/dvo

e.g.

docker run -i -t -p 5900:5900 -e RESOLUTION=1920x1080 -v /home/cwang/data/TUM/rgbd_dataset_freiburg1_360:/root/dataset paopaorobot/dvo

and input :5900 in vnc viewer to connect the desktop


you can run in the desktop:

roslaunch dvo_benchmark benchmark.launch dataset:=[docker_data_path]

e.g.

roslaunch dvo_benchmark benchmark.launch dataset:=/root/dataset/

Maybe you cannot see anything now, the key r need to be pressed. Now you can see the groudtruth. If you want to estimation the trajectory, the key p need to be pressed. Now you can see trajecoties in the window. After all, you will find the result in /root/fuerte_workspace/dvo/dvo_benchmark/output

Publications

The following publications describe the approach:

  • Robust Odometry Estimation for RGB-D Cameras (C. Kerl, J. Sturm, D. Cremers), In Proc. of the IEEE Int. Conf. on Robotics and Automation (ICRA), 2013
  • Real-Time Visual Odometry from Dense RGB-D Images (F. Steinbruecker, J. Sturm, D. Cremers), In Workshop on Live Dense Reconstruction with Moving Cameras at the Intl. Conf. on Computer Vision (ICCV), 2011.

License

The packages dvo_core, dvo_ros, and dvo_benchmark are licensed under the GNU General Public License Version 3 (GPLv3), see http://www.gnu.org/licenses/gpl.html.

The package sophus is licensed under the MIT License, see http://opensource.org/licenses/MIT.

Releases

No releases published

Packages

No packages published

Languages

  • C++ 95.5%
  • CMake 2.3%
  • Python 1.1%
  • Other 1.1%