Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
15,624 datasets
VideoNet is a dataset highlighted at CVPR 2026 for studying domain-specific action recognition and in-context video learning in Vision-Language Models (VLMs). The dataset includes benchmark MP4 files and JSONL files containing question-and-answer pairs. It was uploaded by author 'raivn' and last updated on May 6, 2026.
María-José Romero-Jiménez provides a 1.5 GB collection of ImageJ scripts and models for pixel-based image segmentation of leaf rust. The dataset includes files in MODEL, BSH, ARFF, and IJM formats and was last updated on April 22, 2026. It is shared under a CC-BY-4.0 license.
Pacific, Atlantic, Indian Oceans and Gulf of Mexico surface water measurements from 21 cruises between 1991-07-12 and 2020-04-15. The dataset consists of co-located discrete and underway measurements of fugacity of CO2 (fCO2), alkalinity, and total dissolved inorganic carbon. It was published by the National Oceanic and Atmospheric Administration to infer the temperature dependence of CO2 fugacity in seawater.
NASA's SWOT Medium-accuracy Orbit Ephemeris (MOE) provides daily position and velocity vectors for a satellite's center of mass. The data is organized into 26-hour netCDF-4 files centered on each day in TAI time, with a latency of less than 1.5 days. This dataset is used for forward stream processing in satellite mission analysis.
NASA's SWOT Precise Orbit Ephemeris (POE) dataset provides daily position and velocity vectors for the satellite's center of mass, used in the first SWOT mission reprocessing. Each file spans 26 hours of data centered at 12:00:00 TAI and is delivered in netCDF-4 format with a latency of less than 35 days. The data supports high-precision orbit determination and geophysical analysis.
A computer vision dataset hosted on Hugging Face by vrg-prague, last updated on May 6, 2026. The dataset uses gated access with automatic approval to comply with NeurIPS submission requirements. Accessing the dataset via Hugging Face shares the requester's username and email with the authors.
Sidewalk Management System data tracks inspections and violations for New York City sidewalks. It is published by the New York City Department of Transportation (DOT) and was last updated in April 2026. The dataset includes fields for inspection dates, damage types, and violation statuses.
A civil-infrastructure visual inspection dataset for instance segmentation with 6 defect/condition categories: Algae, Crack, Net-Crack, Crack with Precipitation, Rust, and Spalling. Each sample is either a full-resolution inspection image or a 1024×1024 tile derived from one. The dataset was created by ibm-research and was last updated on Hugging Face in May 2026.
Comprising ablation experiment results for the CRDFNet semantic segmentation model on the MSIDBG remote sensing dataset. The data is stored in an XLS file sized at 5.5 KB. The experiments were conducted by Xin Wang to validate the model's performance on metrics like F1 score, OA, and mIoU.
A collection of ablation experiment results for the CRDFNet semantic segmentation network, tested on the Vaihingen dataset. The network is designed to address challenges in remote sensing imagery such as complex boundary shapes and dense small targets. The data is stored in an XLS file with a size of 5.5 KB.
Comprising ablation experiment results for a semantic segmentation network (CRDFNet) tested on the Potsdam remote sensing dataset. The data is stored in an XLS file sized 5.5 KB. The experiments validated the network's performance on metrics like F1 score, OA, and mIoU.
IJmond Industrial Smoke Segmentation Dataset contains 900 raw images and 2074 cropped images for pixel-level classification. The collection includes 1209 annotated smoke polygons and 1109 cropped images with corresponding pixel-level masks. It was created by Y.C. Hsu and is hosted on figshare.
Images from the Trilateral Monitoring and Assessment Programme (TMAP) for the Wadden Sea. The data is provided by the Wadden Sea Secretariat and published under a Creative Commons Public Domain Mark license via the eu_open_data platform. The last update date is unknown.
668.0 KB of data on dissolved organic matter from the Western North Pacific transect, published by yan chen on figshare in April 2026. The dataset uses optical and molecular proxies to reveal the distribution and transformation of organic matter in this oceanic region.
From August 19, 1977, to January 9, 1979, the HEAO 1 A3 experiment collected X-ray source data using modulation collimators with a 4x4 degree field of view. This catalog contains possible X-ray sources identified by the HEAO 1 A1 instrument. The data is provided as a service by NASA's High Energy Astrophysics Science Archive Research Center (HEASARC).
Demographic characteristics for a cohort of 844 studied patients. The dataset was authored by Atalay Mulu Fentie and is available under a CC-BY-4.0 license. It was last updated on April 21, 2026, and is stored in an XLS file format.
Avinash Bansal's 5.5 KB Excel file, published on figshare in April 2026, compares the effect of individual data augmentation techniques on keypoint detection performance. The dataset likely contains performance metrics for multiple models tested on an augmented dataset, with top-performing results highlighted in bold. Its small size suggests it is a focused summary of experimental results rather than raw training data.
9.5 KB Excel file containing regression coefficients from analyses of points with high adversarial scores derived from a Dynamic Graph Convolutional Neural Network (DGCNN). The dataset, authored by Hanieh Naderi and last updated on April 21, 2026, likely contains significant coefficients shown with three-decimal precision and insignificant ones shown as zero.
NCA Organogram is a dataset published on the eu_open_data platform by the Government Digital Service. It is licensed under CC-BY-4.0 and is available in CSV format. The dataset likely contains information about the organizational structure of the NCA.
2017 surveys collected total sediment metabolism, carbonate, and organic isotope measurements from seabed sediments in two Australian harbours. Sampling was conducted over three weeks by Geoscience Australia, the Australian Institute of Marine Science, and the Northern Territory Government. This data supports a four-year science program from 2014-2018 aimed at creating baseline habitat maps for marine resource management.