7820686
doi
10.48550/arXiv.2304.09704
oai:zenodo.org:7820686
user-eu
Elliot Vincent
LIGM, Ecole des Ponts, Univ Gustave Eiffel, CNRS, France - INRIA and DIENS (ENS-PSL, CNRS, INRIA)
Mathieu Aubry
LIGM, Ecole des Ponts, Univ Gustave Eiffel, CNRS, France
Loic Landrieu
LIGM, Ecole des Ponts, Univ Gustave Eiffel, CNRS, France - LASTIG, Univ Gustave Eiffel, IGN, ENSG
Earth Parser Dataset: A new dataset to train and evaluate parsing methods on large, uncurated aerial LiDAR scans
Romain Loiseau
LIGM, Ecole des Ponts, Univ Gustave Eiffel, CNRS, France - LASTIG, Univ Gustave Eiffel, IGN, ENSG
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
<p>We introduce a new dataset to train and evaluate parsing methods on large, uncurated aerial LiDAR scans. We use data from the French Mapping Agency associated to the LiDAR-HD project. We selected 7 scenes, covering over 7.7km2 and a total of 98 million 3D points, with diverse content and complexity, such as dense habitations, forests, or complex industrial facilities.</p>
<p>You can download sequences individually or use zenodo-get to download all sequences at once:</p>
<p><strong><em>pip install zenodo-get</em></strong></p>
<p><strong><em>zenodo-get 7820686</em></strong></p>
<p>See companion <a href="https://github.com/romainloiseau/EarthParserDataset">github</a> repository and the dedicated <a href="https://imagine.enpc.fr/~loiseaur/learnable-earth-parser">wepage</a> for more information.</p>
<p><strong>Cite as:</strong></p>
<pre><code>@misc{loiseau2023learnable,
title={Learnable Earth Parser: Discovering 3D Prototypes in Aerial Scans},
author={Romain Loiseau and Elliot Vincent and Mathieu Aubry and Loic Landrieu},
year={2023},
eprint={2304.09704},
archivePrefix={arXiv},
primaryClass={cs.CV}
}</code></pre>
<p><strong>Acknowledgements :</strong></p>
<ul>
<li>This work was supported by ANR project READY3D ANR-19-CE23-0007.</li>
<li>The work of MA was partly supported by the European Research Council (ERC project DISCOVER, number 101076028).</li>
<li>The scenes of the Earth Parser Dataset were acquired and annotated by the <strong><a href="https://geoservices.ign.fr/lidarhd">LiDAR-HD</a></strong> project.</li>
<li>We thank Zenodo for hosting the dataset.</li>
</ul>
Zenodo
2023-04-19
info:eu-repo/semantics/other
7820685
user-eu
award_title=Discovering and Analyzing Visual Structures; award_number=101076028; award_identifiers_scheme=url; award_identifiers_identifier=https://cordis.europa.eu/projects/101076028; funder_id=00k4n6c32; funder_name=European Commission;
award_title=Real-Time Analysis of Dynamic LiDAR 3D Point Clouds; award_number=ANR-19-CE23-0007; funder_id=00rbzpz17; funder_name=Agence Nationale de la Recherche;
1692794080.78801
46102065
md5:f9ca767bca408b74ce1d62557e2a8d5a
https://zenodo.org/records/7820686/files/greenhouse.zip
17192197
md5:335fea901003209c922bf69842721a64
https://zenodo.org/records/7820686/files/marina.zip
802444521
md5:66b22c63d636eb6c60db2499b24369d5
https://zenodo.org/records/7820686/files/crop_field.zip
706900820
md5:7c032ddbe7fa8b708b82b231b85ef7f3
https://zenodo.org/records/7820686/files/urban.zip
345791464
md5:fb7677212957aa4ce65639fd1a91a8c5
https://zenodo.org/records/7820686/files/windturbine.zip
2111283787
md5:3bf07fc6a4145f9cfde61e1ffe0e7e07
https://zenodo.org/records/7820686/files/forest.zip
264528804
md5:f702bdb7ecf1f2bf743d2bd72ac89f1d
https://zenodo.org/records/7820686/files/power_plant.zip
public