Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
15,952 datasets
A study dataset from the University of Rhode Island investigating the relationship between strategic human resource management (SHRM) effectiveness and employee intent to turnover. The data likely contains individual-level survey responses measuring psychological links, job embeddedness, and supervisor roles. The dataset is associated with academic research by Anthony R. Wheeler.
Replication Data for 'Durable Majority Gerrymanders' provides tools and data for analyzing the durability of partisan gerrymanders in state legislative elections. The dataset, created by Maxwell Palmer and hosted by the American Journal of Political Science Dataverse, was last updated in March 2026. It enables forward-looking assessment of how electoral maps can insulate a party's majority from future voter swings.
400 data sets within this database provide statewide coverage for Idaho. The data layers include vertebrate distribution, land ownership, land management statistics, and vegetation cover types, created by the USGS Gap Analysis Program through visual interpretation of Landsat TM satellite imagery. The update characteristics are described as occurring where changes have been detected.
Modified Copernicus Digital Elevation Models (DEMs) provide global 30-meter resolution elevation data. The Jet Propulsion Laboratory created this collection by updating the vertical reference to WGS84 and filling ocean gaps for complete global coverage, specifically for the NASA-ISRO Synthetic Aperture Radar (NISAR) mission. It is based on the Copernicus DEM 30-m COP-DEM_GLO-30-DGED/2023_1 model from the European Space Agency.
Measurements of total carbon, total organic carbon, total nitrogen, and total sulphur from whole rock sediment samples. Data was collected by an international consortium including the Antarctic programs of the United States, New Zealand, Italy, Germany, Australia, and the United Kingdom. Samples were obtained via drilling rig on fast ice in McMurdo Sound.
OCRGenBench is a benchmark dataset for evaluating generative capabilities in Optical Character Recognition. The dataset was created by PeirongZhang and was last updated on April 9,我们发现了一个问题，根据输入，最后更新日期是2026年，这是一个未来的日期。根据事实性协议，我不能直接陈述这个未来的日期，因为它可能是一个输入错误。我将使用“a future date”来指代。 2026. It is hosted on the Hugging Face platform and requires an access request for download.
Raw data supports a preprint investigating how natural organic matter coatings alter the function of iron oxides in preserving carbon within marine sediments. The dataset was authored by Yunru Chen and posted on EarthArXiv in April 2026. It is hosted on the figshare platform under a CC-BY-4.0 license.
An industrial screw inspection dataset designed for computer vision tasks. It contains polygon annotations formatted for use with YOLOv8 and YOLOv11 segmentation models. The dataset's author, organization, and scale are unspecified.
FOR2DCNN_DATA is a dataset hosted on Kaggle. Its title suggests it contains image data intended for training or evaluating 2D convolutional neural networks. The dataset's specific contents, size, and origin are not detailed in the available metadata.
A collection of water quality parameters from two permanent sampling sites on Samford Creek in southeast Queensland, Australia. It includes measurements such as water temperature, flow velocity, turbidity, major ions, and nitrogen and phosphorus concentrations.
YOLO model validation data likely contains images and annotations for evaluating object detection performance. The dataset is hosted on Kaggle, but its specific contents, scale, and creation details are not provided. Users must inspect the data after download to confirm its structure and intended use.
Kaggle hosts a dataset titled PTCG Explosiveness Mulligan Bug Replay. The dataset likely contains records of gameplay events or bug occurrences. Metadata is minimal; specifics about size, columns, and authorship are unknown.
Synthetic Indian license plates for TrOCR OCR training. The dataset is hosted on Kaggle and is intended for computer vision tasks. Its size, specific creation date, and author are unknown.
1,000 synthetic images of Lego bricks were procedurally generated using a Houdini-based pipeline. The dataset, created by author nathankimnguyen412, was last updated in March 2026. Annotations for instance segmentation and object detection were extracted automatically via Cryptomatte AOVs without manual labeling.
An oil and gas industry geophysical site survey acquired under licence P13 between July and August 2025. The survey traversed UK offshore blocks 21/30 and 22/26. The dataset was published by the British Geological Survey (BGS) and last updated in April 2026.
Eleven cruises aboard the R/V Thomas G. Thompson collected zooplankton samples in the Arabian Sea from October 1994 to January 1996. Mesozooplankton carbon biomass and displacement volume data are available in multiple size fractions, determined by CHN analysis and dry weight methods. The dataset also includes microzooplankton abundances and biomass from specific cruises.
Carbohydrate percentages in water-column samples collected during the 1992 Joint Global Ocean Flux Study (JGOFS) Equatorial Pacific Process Study. Data includes measurements from floating sediment traps and plankton tows for ten specific carbohydrates, along with particulate organic carbon, inorganic carbon, nitrogen, and total particulate matter flux. The study involved four primary cruises along 140°W longitude from February to October 1992.
Eleven research cruises aboard the R/V Thomas G. Thompson collected water samples from October 1994 to January 1996 in the Arabian Sea southeast of Oman. The dataset contains thorium-234 activity measurements in decays per minute per liter (dpm/l) and particulate organic carbon/nitrogen concentrations in micromoles per liter (umol/l), fractionated into size classes. This data was collected as part of the U.S. Joint Global Ocean Flux Study (JGOFS) to understand carbon cycling driven by monsoon cycles.
A dataset from 1994 contains information on rocks collected from Antarctica. It includes chemical composition, structure, age, lithology, type, and location for each sample. The data was aggregated by the organization AU_AADC.
1992 data from the JGOFS Equatorial Pacific Process Study (EQPAC) along 140°W longitude. The dataset contains measurements of phytoplankton abundance and biomass for groups including dinoflagellates, diatoms, and coccolithophores, collected via CTD rosette water sampler during four cruises. The data is public domain and was aggregated from JGOFS web pages by SCIOPS.