1. Full name: 
Vaihingen Double Layer Dataset 



2. Short name: 
Vaihingen-DL



3. Contents: 
The Vaihingen-DL dataset contains aerial images of Vaihingen village in Germany (000_img.jpg), associated with corresponding digital surface models (DSM) (000_dsm.jpg) and two ground truth images - one for the base (000_base.bmp) and the second - for the occlusion layer (000_occl.bmp). The base layer has 11 classes:
class 00: Road
class 01: Traffic island (asphalt)
class 02: Sidewalk 
class 03: House 
class 04: Grass 
class 05: Agro 
class 06: Water 
class 07: Sealed 
class 08: Traffic island (vegetation) 
class 09: Beach 
class 10: Railway

The occlusion layer has 4 classes:
class 00: Void
class 01: Tree
class 02: Car
class 03: Bridge

In ground truth images, the classes are marked with pixels with the above mentioned values (e.g. classes “Sidewalk” and “Car” have values 4). 



4. Usage:
The Vaihingen-DL dataset can be used to test Image segmentation, feature extraction, classification approaches, etc. Copyright of the images in the dataset fully belongs to their owners. In no event, shall owners be liable for any incidents, or damages caused by the direct or indirect usage of the dataset. The dataset should be only used for non-commercial research and/or educational purposes. 



5. Dataset organizer and ground truth makers: 
M.Sc. Sergey Kosov, University of Siegen, Germany
M.Sc. Alina Maas, Leibniz University Hannover, Germany
et. al.



6. References:
***User shall properly cite at least one of the given below references whenever the dataset is used***

[1] Sergey Kosov, Kimiaki Shirahama and Marcin Grzegorzek, "Labeling of partially occluded regions via the multi-layer CRF", Multimedia Tools and Applications, July 2018.

@article{KosovMTAP2018,
  author = {Kosov, Sergey and Shirahama, Kimiaki and Grzegorzek, Marcin},
  title = {Labeling of partially occluded regions via the multi-layer {CRF}},
  journal = {Multimedia Tools and Applications},
  year = {2018},
  month = {Jul},
  day = {11},
  issn = {1573-7721},
  doi = {10.1007/s11042-018-6298-5},
  url = {https://doi.org/10.1007/s11042-018-6298-5}
}


[2] Sergey Kosov, Pushmeet Kohli, Franz Rottensteiner, and Christian Heipke. A Two-Layer Conditional Random Field for The Classification of Partially Occluded Objects. http://arxiv.org/abs/1307.3043. arXiv:1307.3043 [cs.CV]. 2013.


[3] Sergey Kosov, Franz Rottensteiner, and Christian Heipke. Sequential Gaussian Mixture Models for Two-Level Conditional Random Fields. In proc. of the 35th German Conference on Pattern Recognition (GCPR). Vol. 8142. Lecture Notes in Computer Science. Springer, 2013, pp. 153–163.


[4] Michael Cramer. The DGPF test on digital aerial camera evaluation – overview and test design., In: Photogrammetrie-Fernerkundung-Geoinformation 2, pp. 73–82, 2010.
