Last Modified: January 13, 2020
_____________________________________________________________________

Freiburg Terrains DATASET

@author
         Jannik Zürn
         zuern@informatik.uni-freiburg.de
         http://deepterrain.cs.uni-freiburg.de
_____________________________________________________________________

This package contains the Freiburg Terrains dataset that was introduced in the paper "Self-Supervised Visual Terrain Classification from Unsupervised Acoustic Feature Learning". It contains a total of 24324 images that were collected using a ZED stereo camera using our OBELIX robot platform on the University of Freiburg campus and neighboring areas. It also contains the respective audio recordings obtained with a directional microphone mounted close to the wheels of the robots. We used a highly accurate LiDAR based SLAM system to obtain pose labels. For more details regarding the data collection procedure please refer to the original publication.

_____________________________________________________________________

DESCRIPTION

The full dataset contains the following parts


*	audio 			Contains the time-synchronized audio recordings from the onboard microphone.

*	camera_poses	The file `ri_slam_solution_final.gm2dl` contains the robot pose graph in format `x, y, z, qw, qx, qy, qz`, and the file `velodyne_timestamps.txt` comtains the UNIX timestamp for each pose.

*	rgb 			Contains the RGB images obtained from the ZED-camera and a file with the UNIX timestamp of each image.



_____________________________________________________________________

CITATION

If you use this dataset or report results on it, please consider citing our paper below:

J. Zürn, W. Burgard, and A. Valada, "Self-Supervised Visual Terrain Classification from Unsupervised Acoustic Feature Learning", arXiv Preprint arXiv:1912.03227, 2019. 

@article{zurn2019self,
  title={Self-Supervised Visual Terrain Classification from Unsupervised Acoustic Feature Learning},
  author={Z{\"u}rn, Jannik and Burgard, Wolfram and Valada, Abhinav},
  journal={arXiv preprint arXiv:1912.03227},
  year={2019}
}

_____________________________________________________________________


==========================================================
Changelog
==========================================================

- v0.1 Initial public release