Login
- Titel:
GlobalBuildingMap
- Dokumenttyp:
- Forschungsdaten
- Veröffentlichungsdatum:
- 11.04.2025
- Verantwortlich:
- Zhu, Xiaoxiang
- Autorinnen / Autoren:
- Zhu, Xiaoxiang; Li, Qingyu; Shi, Yilei; Wang, Yuanyuan; Stewart, Adam J.; Prexl, Jonathan
- Institutionszugehörigkeit:
- TUM
- Herausgeber:
- TUM
- Identifikator:
- doi:10.14459/2024mp1764505.001
- Konzept-DOI:
- doi:10.14459/2024mp1764505
- Enddatum der Datenerzeugung:
- 01.11.2024
- Fachgebiet:
- DAT Datenverarbeitung, Informatik; GEO Geowissenschaften
- Quellen der Daten:
- Abbildungen von Objekten / image of objects
- Datentyp:
- Bilder / images
- Methode der Datenerhebung:
- The GlobalBuildingMap is generated by applying an ensemble of deep neural networks on nearly 800,000 satellite images of about 3m resolution. The deep neural networks were trained with manually inspected training samples generated from OpenStreetMap. Evaluation of GlobalBuildingMap were conducted on 34 unseen cities. It reaches an overall F1 score of 0.54, comparing to 0.47 for Microsoft building footprint and 0.20 for Google building footprint.
- Beschreibung:
- The GlobalBuildingMap (GBM) dataset provides the highest resolution and highest accuracy building footprint map on a global scale ever created. GBM was generated by training and applying modern deep neural networks on nearly 800,000 satellite images. The dataset is stored in 5 by 5 degree tiles in geotiff format. This is the first version of the dataset, which contains the African continent.
- Technische Hinweise:
- View and download (8,5 GB total, 170 Files)
The data server also offers downloads with FTP
The data server also offers downloads with rsync (password m1764505):
rsync rsync://m1764505@dataserv.ub.tum.de/m1764505/ - Sprache:
- en
- Rechte:
- by, http://creativecommons.org/licenses/by/4.0
- Horizon 2020:
- ERC-StG-714087
BibTeX