Skyscapes -- Fine-Grained Semantic Understanding of Aerial Scenes
Azimi, Seyed Majid and Henry, Corentin and Sommer, Lars and Schumann, Arne and Vig, Eleonora
German Aerospace Center (DLR: Deutsches Zentrum für Luft- und Raumfahrt e.V.)
The work has been accepted in ICCV19 conference.
You can download the dataset from here https://eoc-datasets.dlr.de
NOTE: we are currently setting up an online evaluation server. As soon as the server is online, you will be notifed in this repo. Thanks for your interest in SkyScapes.
Please contact us using this link: https://www.dlr.de/eoc/en/desktopdefault.aspx/tabid-12760/22294_read-58694 to receive the dataset.
TUM Profile: Technical University of Munich (TUM): https://www.bgu.tum.de/lmf/azimi/
We appreciate the help of Torno AI GmbH https://torno.tech for its assistance with the annotation process to achieve very high-quality, inexpensive, safe and fast annotation data.
If you use SkyScapes dataset in your research, please star the project and consider citing:
@inproceedings{azimi2019skyscapes,
title={SkyScapes - Fine-Grained Semantic Understanding of Aerial Scenes},
author={Azimi, Seyed Majid and Henry, Corentin and Sommer, Lars and Schumann, Arne and Vig, Eleonora},
booktitle={Proceedings of the IEEE international conference on computer vision},
year={2019}
}
@article{azimi2018aerial,
title={Aerial LaneNet: Lane-marking semantic segmentation in aerial imagery using wavelet-enhanced cost-sensitive symmetric fully convolutional neural networks},
author={Azimi, Seyed Majid and Fischer, Peter and K{\"o}rner, Marco and Reinartz, Peter},
journal={IEEE Transactions on Geoscience and Remote Sensing},
volume={57},
number={5},
pages={2920--2938},
year={2018},
publisher={IEEE}
}
We are looking for Master and PhD students with Computer Vision knowledge and Deep Learning experience to work in the field of remote sensing in German Aerospace Center (DLR) in Muenchener Str. 20, 82234 Wessling, Germany.