[robotics-worldwide] [software] ORB-SLAM: a Versatile and Accurate Monocular SLAM

Raúl Mur Artal raulmur at unizar.es
Tue Mar 10 02:58:20 PDT 2015


Dear community,

We have just released the source code of ORB-SLAM.
https://github.com/raulmur/ORB_SLAM

ORB-SLAM is a new feature-based Monocular SLAM very precise and robust.
It works successfully in challenging public datasets as NewCollege, 
KITTI and TUM RGB-D (using only RGB).
Typical RMSE for the keyframe poses is around 1cm in the TUM RGB-D 
benchmark and of few meters in the KITTI dataset.

Videos, related publications and the source code can be found at:
http://webdiis.unizar.es/~raulmur/orbslam/

Our ORB-SLAM paper is currently under review in IEEE Transactions on 
Robotics, but a preprint is available in ArXiv:
http://arxiv.org/abs/1502.00956

Best regards,
Raúl Mur-Artal
J.M.M Montiel
Juan D. Tardós

-- 
Raúl Mur Artal

http://webdiis.unizar.es/~raulmur/
raulmur at unizar.es

Universidad de Zaragoza
Departamento de Informática e Ingeniería de Sistemas
Escuela de Ingeniería y Arquitectura
Edificio Ada Byron, L1.01



More information about the robotics-worldwide mailing list