EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LAB-Net: Lidar and aerial image-based building segmentation using U-Nets

Satheshkumar Kaliyugarasan, Alexander Selvikvåg Lundervold

Nordic Machine Intelligence · 2023

DOI: 10.5617/nmi.10163 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Nordic Machine Intelligence, 2023. Verfasst von Satheshkumar Kaliyugarasan und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We describe our approach in the 2022 NORA MapAI competition. The objective of the competition was to construct methods that were able to detect and segment buildings from aerial imaging and laser data. There were two subtasks: (1) building segmentation from aerial imaging; (2) building segmentation from lidar data, optionally combined with aerial images. We trained multiple dynamic U-Net models with self-attention layers. For Task 1, we used a ResNet34-based encoder pre-trained on the ImageNet challenge dataset and further pre-trained the U-Net on another similar aerial image dataset. For Task 2, we adapted the dynamic U-Net to deal with multispectral data. Our ensembles placed us in second place, with the top score on Task 1. The complete source code for reproducing our results is available at https://github.com/HVL-ML/LAB-Net.

Zitieren

APA

Kaliyugarasan, S., & Lundervold, A. S. (2023). LAB-Net: Lidar and aerial image-based building segmentation using U-Nets. Nordic Machine Intelligence. https://doi.org/10.5617/nmi.10163

Vancouver

Kaliyugarasan S, Lundervold AS. LAB-Net: Lidar and aerial image-based building segmentation using U-Nets. Nordic Machine Intelligence. 2023. doi:10.5617/nmi.10163

BibTeX
@article{Kaliyugarasan2023,
  author = {Kaliyugarasan, Satheshkumar and Lundervold, Alexander Selvikvåg},
  title = {LAB-Net: Lidar and aerial image-based building segmentation using U-Nets},
  journal = {Nordic Machine Intelligence},
  year = {2023},
  doi = {10.5617/nmi.10163}
}
Kostenlos starten →