Loading...
Loading...
Self-driving perception, LiDAR/camera fusion, trajectory prediction, drone perception, robot manipulation
2,023 datasets
Florida Keys coastal and seafloor elevation data collected by Quantum Spatial, Inc. for NOAA after Hurricane Irma. The dataset includes 70 digital elevation model tiles at 1-meter resolution, covering 285,529 acres from Miami south to the Marquesas Keys, with acquisition dates from November 2018 to March 2019. Data points are classified into categories including ground, bathymetric bottom, water surface, and submerged objects.
NOAA Chesapeake Bay MD1804 Option 1 West 4 topobathymetric lidar data were collected by NV5 Geospatial, Inc. using a Riegl VQ-880-GH system. The dataset covers approximately 492.7 square kilometers of the Chesapeake Bay area near Solomons, California, and other Maryland census designated places. It comprises 2,593 LAS tiles, each 500 m x 500 m, collected over nine missions from March to April 2019.
NOAA NGS topobathy lidar data covering approximately 942 square miles of the Texas Gulf coast from Galveston to Corpus Christi. The data were collected by Leading Edge Geomatics using a Leica Chiroptera II sensor between October 22, 2018 and May 31, 2019. The delivery consists of 3,813 classified LAS tiles in five blocks.
Profiles of water vapor mixing ratio and aerosol scattering ratio collected during the CAMEX3 campaign. The data were gathered using Raman Lidar systems that illuminate the atmosphere with pulsed lasers and measure wavelength-shifted molecular returns. The dataset was last updated by SCIOPS in September 1998.
Topobathymetric lidar data covering approximately 311 square miles of the Tampa Bay inlet, including Old Tampa Bay and Hillsborough Bay. The data were collected by Leading Edge Geomatics using a RIEGL VQ880-G system between November 22 and December 20, 2019. The National Oceanic and Atmospheric Administration (NOAA) National Geodetic Survey (NGS) is the authoritative source.
Kaggle hosts a dataset titled 'dronemodel'. The dataset likely contains images or specifications related to drone models. Its content and scale require verification after download.
Islamic Hadith corpus is a dataset hosted on Kaggle. The title suggests it contains collections of Hadith, which are reports of the sayings and actions of the Prophet Muhammad. The dataset's specific content, size, and provenance require verification after download.
VisDrone2019-MOT-Custom is a dataset derived from the VisDrone2019 benchmark, likely containing sequences for multi-object tracking tasks. The dataset is hosted on Kaggle, but specific details about its size, annotations, and modifications are not provided in the available metadata. Its content and structure must be verified after download.
A multimodal autonomous driving dataset includes data from a 128-beam Velodyne Alpha-Prime lidar, a 5MP Blackfly camera, a 360-degree Navtech radar, and post-processed Applanix POS LV GNSS data. It was collected by ASRL over the course of a year in various weather conditions including sun, rain, and snow. The intended purpose is to enable benchmarking of long-term all-weather odometry and metric localization.
V2X R is a dataset hosted on HuggingFace by author hx24. The dataset was last updated on April 24, 2026. Its specific content, size, and structure are not detailed in the available metadata.
Lidards is a dataset published on Kaggle. The title suggests it contains Light Detection and Ranging (LiDAR) data, likely consisting of 3D point clouds. Metadata is minimal; actual content, scale, and specific features require verification after download.
Person-UAV-Manual is a dataset hosted on Kaggle. The title suggests it contains data related to person detection from unmanned aerial vehicles (UAVs), likely involving manual annotations. No further metadata on size, columns, or origin is available.
Kitware provides a large-scale multi-sensor dataset for autonomous, off-road navigation. It features simultaneously collected imagery from multispectral, hyperspectral, polarimetric, and broadband sensors, spanning wavelengths from visible to long-wave infrared, with aligned LIDAR data for ground-truth shape. Camera calibrations, LiDAR registrations, and traversability annotations are available for a subset.
New Jersey elevation data consists of 18 discrete LiDAR collection projects covering overlapping subsections of the state. Projects were conducted over several years with differing technical specifications and include deliverables such as classified point clouds, digital elevation models, and contour lines. The data adheres to Federal Geographic Data Committee metadata standards.
VisDrone2019 test set results compare detection models for drone-captured imagery. The dataset contains performance metrics in a 5.5 KB XLS file, authored by Linfeng Jia and updated in March 2026.
5.5 KB XLS file compares object detection model performance on the HIT-UAV test set. The data, authored by Linfeng Jia, was last updated in March 2026 and is licensed for open use under CC BY 4.0.
Targeted ablation results of cross-branch interaction in the DPD head on the HIT-UAV dataset. The 5.5 KB XLS file contains experimental metrics from model ablation studies. It was authored by Linfeng Jia and last updated in March 2026.
A multimodal dataset fusing hyperspectral and LiDAR data for urban land cover classification. The dataset was collected over Houston, Texas, in 2013. The specific author, organization, and data volume are not provided in the available metadata.
Multi-modal sensor data was collected from fully autonomous Indy race cars operating at speeds up to 170 mph (273 kph). Six teams contributed data from the 2021-22 Indy Autonomous Challenge, covering 11 racing scenarios across two tracks. The data is organized and released in both ROS2 and nuScenes formats.
Northern Ireland's entire coastline, including intertidal areas and approximately 200 meters landward of the high water mark, is covered by this orthophotography. The geometrically corrected aerial imagery was captured as part of the 2021 NI 3D Coastal Survey with a 10cm resolution. The dataset is provided by OpenDataNI under the OGL-UK-3.0 license.