Roof Segmentation based on Deep Neural Networks

Regina Pohle-Fröhlich, Aaron Bohm, Peer Ueberholz, Maximilian Korb, Steffen Goebbels

2019

Abstract

The given paper investigates deep neural networks (DNNs) for segmentation of roof regions in the context of 3D building reconstruction. Point clouds as well as derived depth and density images are used as input data. For 3D building model generation we follow a data driven approach, because it allows the reconstruction of roofs with more complex geometries than model driven methods. For this purpose, we need a preprocessing step that segments roof regions of buildings according to the orientation of their slopes. In this paper, we test three different DNNs and compare results with standard methods using thresholds either on gradients of 2D height maps or on point normals. For the application of a U-Net, we transform point clouds to structured 2D height maps, too. PointNet and PointNet++ directly accept unstructured point clouds as input data. Compared to classical gradient and normal based threshold methods, our experiments with U-Net and PointNet++ lead to better segmentation of roof structures.

Paper Citation

in Harvard Style

Pohle-Fröhlich R., Bohm A., Ueberholz P., Korb M. and Goebbels S. (2019). Roof Segmentation based on Deep Neural Networks. In Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2019) - Volume 4: VISAPP; ISBN 978-989-758-354-4, SciTePress, pages 326-333. DOI: 10.5220/0007343803260333

in Bibtex Style

@conference{visapp19,
author={Regina Pohle-Fröhlich and Aaron Bohm and Peer Ueberholz and Maximilian Korb and Steffen Goebbels},
title={Roof Segmentation based on Deep Neural Networks},
booktitle={Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2019) - Volume 4: VISAPP},
year={2019},
pages={326-333},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0007343803260333},
isbn={978-989-758-354-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2019) - Volume 4: VISAPP
TI - Roof Segmentation based on Deep Neural Networks
SN - 978-989-758-354-4
AU - Pohle-Fröhlich R.
AU - Bohm A.
AU - Ueberholz P.
AU - Korb M.
AU - Goebbels S.
PY - 2019
SP - 326
EP - 333
DO - 10.5220/0007343803260333
PB - SciTePress