Loading...
Loading...
Self-driving perception, LiDAR/camera fusion, trajectory prediction, drone perception, robot manipulation
2,003 datasets
State-level data for Pennsylvania (2008), Delaware (2014), and Maryland (2013) provides 1-meter resolution tree canopy cover for the Northeast USA. The data were created using a rules-based expert system that integrated leaf-on LiDAR and imagery, exploiting spectral, height, and spatial information. Additional states are planned to be added as processing is completed.
2012 estimates of forest aboveground biomass for the Paragominas municipality in Para, Brazil. The dataset was produced by NASA using gradient boosting machine models to assimilate ground inventory plot data with LiDAR and PALSAR remote sensing metrics. It provides spatially explicit predictions of biomass and uncertainty for three focal study areas and the entire municipality.
Transparency International's Corruption Perceptions Index scores countries from 0 (highly corrupt) to 100 (very clean). The dataset includes scores and rankings based on assessments by experts and business executives. It was last updated on 2026-05-21.
Songdo Vision is a high-resolution aerial vehicle-detection dataset of 5,419 annotated 4K (3840 × 2160) bird's-eye view RGB drone frames. It contains approximately 272,000 vehicle instances across four classes (car, bus, truck, motorcycle), captured during a multi-drone urban traffic monitoring experiment over Songdo. The dataset was created by author rfonod and was last updated on 2026-06-25.
High-resolution maps estimate aboveground biomass for four distinct US forest sites in 2011. The 11 GeoTIFF files provide carbon stock estimates at 20-50 meter resolution, generated by combining field inventory data with LiDAR remote sensing and modeling. Uncertainty estimates are included for the Maryland site using two different methodologies.
Two lidar systems, the Semi Autonomous Tropospheric Aerosol Lidar and the Tropospheric Water Vapor Lidar, collected imagery during the GPM Cold-season Precipitation Experiment from January to March 2012 in Canada. This dataset provides quicklook imagery to address shortcomings in GPM snowfall retrieval algorithms by capturing microphysical properties and remote sensing observations of precipitating snow. The system operated semi-autonomously, shutting down automatically during rain events.
A modified Sea Horse Optimization algorithm (moSHO) designed for threat-aware unmanned aerial vehicle path planning. The dataset likely contains results from evaluating moSHO on 23 benchmark functions and UAV path planning experiments. Amir Seyyedabbasi authored this dataset, which was last updated on May 11, 2026.
A 5.7 KB CSV file presents a modified Sea Horse Optimization algorithm (moSHO) for threat-aware unmanned aerial vehicle path planning. The dataset, authored by Amir Seyyedabbasi and last updated in May 2026, describes an algorithm that integrates three cooperative strategies to improve exploration and avoid local optima. It includes evaluation results from 23 benchmark functions and experiments on a UAV path planning model under threat environments.
Coastal LiDAR data provides a 0.25-meter resolution contour model for the City of Hobart. This dataset, sourced from the City of Hobart Open Data platform, is available in multiple geospatial formats including GeoJSON and KML. Its cross-platform presence indicates its established use for terrain analysis in the region.
Lidar data provides 3D point clouds with X, Y, Z coordinates representing precise surface reflections from airborne laser surveys. These datasets result from intergovernmental collaborations with Quebec ministries, federal, and municipal sectors, enabling detailed terrain and surface models. Data availability is indicated on a download map, with coverage expanding as new surveys are completed.
SpatialUAV is a diagnostic benchmark containing 4,331 curated visual question-answering instances for evaluating spatial intelligence in low-altitude unmanned aerial vehicle scenarios. The dataset covers 14 task types, including semantic discrimination, spatial relations, and motion understanding. It was created by Hyu-Zhang and last updated on 2026-06-29.
Multi-taxon data from temporary ponds supports a comparative analysis of metacommunity approaches. The dataset contains environmental, spatial, and temporal variables alongside species matrices for phytoplankton, microinvertebrates, macroinvertebrates, and amphibians. It was created by Ángel Gálvez and colleagues for a study published on paperswithcode.
The ISC2010_STRUCTURE1_R table provides statistical summaries for the Structure1 metric, representing vegetation cover in two height categories (shrubs 1.5m-5m and trees >5m) within a 40m riparian zone. The Department of Environment and Primary Industries (DEPI) developed this data using remote sensing from a 2010-13 state-wide mapping project, specifically 15cm aerial photography and multi-pulse LiDAR. This data is designed to join with river centerline feature classes for assessing stream condition in Victoria.
Structure2 represents vertical vegetation layering across six height intervals from 1.5m to over 25m. The Department of Environment and Primary Industries developed this methodology using remote sensing data, specifically 15cm aerial photography and multi-pulse LiDAR, to assess river condition in Victoria. This statistical summary table is designed to join to a river centerline feature class and is part of a state-wide mapping project conducted from 2010 to 2013.
The ISC2010_FRAGMENTATION_R table is a statistical summary for the Fragmentation Metric at the Reach level, designed to join to the ISC2010_RIVER_CENTRELINES_R feature class. The Department of Environment and Primary Industries (DEPI) developed this data using remote sensing from a 2010-13 state-wide mapping project. Remote sensing data included 15cm true colour and infra-red aerial photography and four-return multi-pulse LiDAR data.
The ISC2010_WATER_BODIES_R table is a statistical summary for the Water Bodies Metric at the Reach level, designed to join to a river centerlines feature class. River condition in Victoria was assessed using the Index of Stream Condition (ISC), with the Department of Environment and Primary Industries (DEPI) developing a methodology using remote sensing data like LiDAR and aerial photography. A state-wide mapping project from 2010-13 derived metrics for Physical Form and Riparian Vegetation.
The ISC2010_WATER_BODIES_S table provides statistical summaries for the Water Bodies Metric at 100-meter sections. It was created by the Department of Environment and Primary Industries (DEPI) using remote sensing data, specifically 15cm aerial photography and multi-pulse LiDAR, collected during a state-wide mapping project from 2010 to 2013. The table is designed to join to a river centerlines feature class for analysis.
The ISC_STREAMBED_WIDTH_S table provides statistical summaries for the Streambed Width metric at 100-meter river sections. The Department of Environment and Primary Industries developed this data using remote sensing from 15cm aerial photography and multi-pulse LiDAR collected during a state-wide mapping project from 2010 to 2013. It is part of a broader effort to assess the Physical Form and Riparian Vegetation components of river health in Victoria.
Polygon features representing non-vegetated bank faces along Victorian rivers, mapped from 2010-2013 remote sensing data. The Department of Environment and Primary Industries (DEPI) developed the methodology using 15cm aerial photography and multi-pulse LiDAR. Areas of bare ground smaller than 5 square meters were excluded to manage data volume.
A 2010-2013 state-wide mapping project in Victoria, Australia, produced these geospatial polygon features representing areas of stream beds that contained water at the time of survey. The Department of Environment and Primary Industries (DEPI) developed the methodology using 15cm aerial photography and LiDAR data to assess Physical Form and Riparian Vegetation metrics for the Index of Stream Condition. The dataset excludes water body features smaller than 5 square meters to manage data volume.