Open Access Benchmarks:
In order to benefit the researchers in the ICA committee, we have published the following open access labeled data sets (or benchmarks) in our Commission Website to promote the use of deep learning algorithms for sensor-driven mapping research and development:
- University of Waterloo releases Waterloo Building Benchmark, https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/EXRA2V
- University of Waterloo makes Toronto-3D MLS Benchmark available, https://github.com/WeikaiTan/Toronto-3D
- University of Waterloo releases OpenGF Ground Filtering Benchmark, https://github.com/Nathan-UW/OpenGF
- University of Dayton releases DALES ALS Benchmark, https://udayton.edu/engineering/research/centers/vision_lab/research/was_data_analysis_and_processing/dale.php
- São Paulo City Hall releases ALS data, http://geosampa.prefeitura.sp.gov.br/PaginasPublicas/_SBC.aspx
- Scale AI releases PandaSet dataset, https://scale.com/open-datasets/pandaset
- Road and Building Detection Datasets, https://www.cs.toronto.edu/~vmnih/data/
- Inria Aerial Image Labeling Benchmark, https://project.inria.fr/aerialimagelabeling/
- SpaceNet on Amazon Web Services, https://registry.opendata.aws/
- Wuhan University releases WHU-TLS Benchmark, https://3s.whu.edu.cn/ybs/en/benchmark.htm
