Loading...
Loading...
Self-driving perception, LiDAR/camera fusion, trajectory prediction, drone perception, robot manipulation
2,015 datasets
ELDOR is a benchmark dataset for detecting illegal gold mining activities in the Amazon Rainforest. The dataset consists of drone-captured orthomosaic imagery and corresponding labels, organized into train, validation, and test splits. It was created by IRSC and documented in a 2026 research paper.
OWLETS2_Surface_Data_1 contains in-situ and remote sensing measurements from the NASA-funded Ozone Water-Land Environmental Transition Study (OWLETS-2) field campaign. The dataset captures ozone concentrations, pollutant gradients, and meteorological variables across land-water boundaries in the Chesapeake Bay region. It was collected from three primary research sites to study fundamental atmospheric processes at the coastal interface.
NASA's Ozone Water-Land Environmental Transition Study (OWLETS-2) collected ozone and nitrogen dioxide data from June 6 to July 6, 2018 in the upper Chesapeake Bay region. The campaign used Pandora spectrometers at ground sites and on a research vessel to characterize pollutant gradients across the land-water interface. This dataset addresses a known measurement gap in coastal air quality research.
NASA's Ozone Water-Land Environmental Transition Study (OWLETS-1) deployed a unique combination of aircraft, ozone lidars, drones, and ship-based measurements to examine coastal air quality. The campaign collected synchronous vertical profiles of pollutants and meteorology over the Chesapeake Bay from July 5 to August 3, 2017, focusing on the land-water interface. This analysis model data is intended to characterize gradients and fundamental processes affecting ozone concentrations in coastal environments.
June 6 to July 6, 2018, the Ozone Water-Land Environmental Transition Study 2 (OWLETS-2) collected aerosol data at ground sites in the upper Chesapeake Bay region. The campaign used ceilometers and PILS-IC instruments to measure aerosol backscatter and composition, focusing on pollutant gradients across the land-water interface. This dataset is part of a broader NASA SIF-funded effort to understand coastal air quality processes.
NVIDIA provides a sample scene dataset for the AlpaDreams autonomous vehicle world model. The dataset's use is governed by the NVIDIA Autonomous Vehicle Dataset License Agreement. It was last updated on June 1, 2026.
FK200930 Northern Great Barrier Reef Bathymetry Acquisition contains fifteen geotiff files of processed multibeam sonar data from a 47-day survey aboard the RV Falkor in 2020. The data, published by Geoscience Australia via the Australian Ocean Data Network, provides 4m to 32m resolution bathymetry of the seabed. It covers the Cape York Peninsula region, Swain slide landslide, and reefs in the eastern Coral Sea Marine Park.
Marie-Josiane Ntsama Essomba published a dataset on figshare in March 2026. It contains baseline characteristics for 108 community-dwelling older adults aged 60 and above from Cameroon. The data was collected to analyze the relationship between probable sarcopenia and domains of intrinsic capacity.
750 valid drone flight records for medical logistics transport between two campuses of Deyang People's Hospital from April 8 to July 30, 2024. It compares drone transport time per unit distance (1.64 ± 0.14 min/km) against road traffic times measured by three navigation apps at nine different time points.
25 last echoes per square meter on average were collected for a 20 km² survey on September 9, 2020. The Digital Surface Model (DSM) was commissioned by the Luing History Group and surveyed by Fugro, covering the Slate Islands off Scotland's west coast. The data offers complete coverage of Luing and Torsa, with partial scans of Fladda, Belnahua, Seil, and Shuna.
11,845 km² of airborne LiDAR data collected across 10 sites in Scotland between March 2011 and May 2012. The Scottish Government, SEPA, and Scottish Water commissioned this dataset for flood risk management, and it has been used for archaeology and orienteering. This specific dataset is the Digital Surface Model (DSM) derived from the point cloud data.
3,516 km² of airborne LiDAR data collected for 66 sites in Scotland between November 29, 2012 and April 18, 2014. The Scottish Public Sector LiDAR (Phase II) dataset was commissioned by the Scottish Government, SEPA, sportscotland, and 13 local authorities under the Flood Risk Management Act (2009). This Digital Surface Model (DSM) is derived from point cloud data and has been used for flood management, archaeology, and orienteering.
VisDro_Title is a hybrid version of the VisDrone dataset. The dataset likely contains images for object detection tasks, with 30% of the images being tiled to increase sample diversity. The author, organization, and specific data size are unknown.
Quranic Recitation Dataset (Word-by-Word Sync) contains high-quality audio recitations of the Holy Quran. The dataset includes recitations from 34 world-renowned reciters across all 114 Surahs, synchronized with millisecond-accurate word timestamps. It was created by zaibihassan and last updated on 2026-05-17.
SPURS-2 Saildrone data provides near-surface oceanographic and atmospheric measurements collected by unmanned surface vehicles during a NASA field campaign. The dataset captures conditions in the eastern tropical Pacific near 10°N, 125°W, focusing on salinity dynamics in a rainfall-dominated region. Data files are in netCDF format and are CF/ACDD/NCEI compliant, containing platform telemetry and sensor readings.
A UAV-derived Digital Terrain Model for the Osoyoos Larose Vineyard in British Columbia's Okanagan Valley. The dataset maps terrain-driven cold-air pooling susceptibility using Topographic Position Index and flow accumulation analysis. It was created by Clarissa Gutierrez and last updated in May 2026.
Multispectral imagery and GIS data from an Unmanned Aerial Vehicle (UAV) were used to identify vineyard rows and segment canopies in the Okanagan region. The dataset includes high-resolution imagery with a 3cm ground pixel size, processed vegetation indices (NDRE, GNDVI), and a canopy height model. Author Kelvin Boateng contributed this dataset to the Borealis Harvested Dataverse, with a last update recorded on 2026-05-02.
18.7 GB of multi-modal roadside perception data collected under diverse environmental conditions. The dataset integrates cameras, various LiDAR sensors, and 4D millimeter-wave radar from seven locations along a 600-meter road segment. Author Bowen Wang published this dataset on figshare in April 2026.
A detailed hydrographic network mapping all visible flow channels across 12 MRCs in Montérégie, Quebec. The project was photo-interpreted using orthophotos from 2018-2020 and LiDAR-derived shaded relief from 2008-2020, funded by the Montérégie FRR program.
A 2019 lidar survey commissioned by the Scottish Environment Protection Agency (SEPA) and flown by Bluesky International Ltd. The 25 cm resolution Digital Surface Model (DSM) covers approximately 200 km² of main settlement areas, including Stornoway, to improve topographic data for flood risk modelling. Data was captured at 16 points per square metre and processed using the OSGM15 geoid model and OSTN15 transformation.