Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
15,790 datasets
Meta-Album's Boats dataset contains 138,367 square images of 26 different maritime vessel classes, resized to 128x128 pixels. It is a preprocessed version of the MARVEL dataset, created by Erhan Gundogdu et al. and released as part of the Meta-Album collection in March 2022. The images were processed by duplicating borders to create squares without cropping, as boats occupy most of the frame.
The PanNuke dataset contains 5,530 images of 19 different human tissue types, transformed from a nuclei segmentation task into a tissue classification task. The images are 128x128 pixels and have undergone stain normalization to reduce bias. It was created by researchers including Jevgenij Gamper and adapted for the Meta-Album benchmark by Romain Mussard in March 2022.
An exploratory pilot study protocol investigating the behavioral and neurophysiological response of individuals with Parkinson's Disease to two types of rhythmic auditory stimulation. The study aims to record local field potentials from implanted deep brain stimulation devices and gait parameters. The protocol was authored by Kyurim Kang and last updated on March 18, 2026.
Concentrations of gas-phase organic and inorganic analytes measured by the Caltech Chemical Ionization Mass Spectrometer (CIT-CIMS) aboard the NASA DC-8 aircraft during four ATom campaigns. The instrument employs CF3O-ion chemistry with two independent mass spectrometers for sensitive measurements. The dataset includes specific measurements for hydrogen peroxide, hydrogen cyanide, nitric acid, methyl hydrogen peroxide, peroxyacetic acid, peroxynitric acid, and sulfur dioxide.
Source data for all graphic figures in the 2026 manuscript 'Pancreatic α-cells are functionally heterogeneous and sex-dependently regulated by neighboring endocrine cells'. The dataset is provided by author Ya-Chi Huang under a CC-BY-4.0 license and was last updated in April 2026. It is a small dataset, 284.3 KB in size, stored in an XLSX file.
A source of Human Development Index (HDI) and Multidimensional Poverty Index (MPI) metrics for Uganda, produced by the UNDP Human Development Reports Office. It tracks progress across health, education, and living standards, incorporating subnational poverty data from the 2019 Global MPI report.
The UNDP Human Development Reports Office (HDRO) maintains this dataset of human development and multidimensional poverty indicators for Lao PDR. It tracks progress across health, education, and standard of living dimensions using standardized metrics like the HDI and the 2019 Global Multidimensional Poverty Index (MPI). The data is intended to facilitate policy discussions by comparing social outcomes against economic growth metrics like GNI per capita.
Human development and multidimensional poverty indicators for the Democratic Republic of the Congo, produced by the UNDP Human Development Reports Office. The data covers dimensions of health, education, and standard of living, including subnational poverty metrics from the 2019 Global MPI report.
A 23.9 KB dataset from figshare, last updated March 2026, details genetic analysis of the fungus Trichoderma reesei. It was created by Yuhang Luo to identify the Tr80437 gene as a key factor for resistance to the inhibitor furfural and for enhanced cellulase production. The data likely contains results from bulked segregant analysis and cross-species screening with Neurospora crassa.
NARSTO_EPA_SS_FRESNO_PM25_OC_EC contains particulate matter measurements from the EPA's Fresno Supersite. The dataset includes hourly measurements of PM2.5 organic and elemental carbon from two continuous monitors operated from January 13, 2000 to March 31, 2005. It was produced under the EPA PM Supersites Program, with data archived by the National Aeronautics and Space Administration.
492 design university students were surveyed to explore the link between academic stress and dependency on AI-generated design tools. The data, published under CC-BY-4.0 by Dan Wang in March 2026, supports analysis of chain mediation effects involving procrastination and help-seeking behavior. It is a 67.5 KB supplementary file from a study extending the I-PACE model to design education.
SUDHIR KUMAR SANGULA's dataset contains annotated knee X-ray images prepared for training YOLO-based deep learning models. The collection is hosted on figshare and was last updated in March 2026. The dataset is relatively small, with a total size of 2.9 MB.
609 Mexican women, including 169 with osteoporosis and 205 with hip fracture, were genotyped for variants in seven genes related to bone metabolism. Antonio Miranda-Duarte published this case-control study on figshare in March 2026. The data includes odds ratios for specific genotypes and results from multifactor dimensionality reduction analysis for gene-gene and gene-environment interactions.
A 2026 case-control study of 609 Mexican women (169 with osteoporosis, 205 with hip fracture, 235 controls) by Antonio Miranda-Duarte. It analyzes single-nucleotide variants in genes like ESR1, RANK, IL6, and their interactions with environmental factors to assess osteoporosis and hip fracture risk using real-time PCR and multifactor dimensionality reduction.
A high-resolution bathymetry surface created from a contracted national reference survey between Gantheaume Point and Talboys Rock, Broome, Western Australia. The survey was acquired for the Australian Hydrographic Office over two days in September 2020 using a Kongsberg EM 2040D multibeam echosounder. The processed data is provided as a 0.5m resolution, 32-bit floating point GeoTIFF grid in MSL, LAT, and Ellipsoid vertical datums.
The CYGNSS Level 1 Calibrated Raw IF Version 1.0 dataset contains the first release of calibrated raw intermediate frequency data from the CYGNSS mission. CYGNSS is a NASA Earth System Science Pathfinder Mission launched on 15 December 2016, consisting of eight micro-satellites providing nearly gap-free Earth coverage. The dataset includes signal coherence detectors, scattered signal strength products, and high-resolution Delay Doppler Maps.
A cross-sectional study of 130 first-episode, drug-naïve MDD patients and 122 healthy controls, investigating relationships between childhood maltreatment and structural brain network organization. The dataset contains connectivity measures for rich club and feeder connections derived from diffusion imaging data. The work was authored by Junhua Huang and last updated on March 18, 2026.
130 first-episode, drug-naïve MDD patients and 122 healthy controls were recruited for a cross-sectional study. The dataset contains connectivity measures for structural brain networks derived from diffusion imaging data, comparing rich club and feeder connections across groups. It was authored by Junhua Huang and last updated on March 18, 2026.
Junhua Huang's research dataset, published on figshare in March 2026, contains results from a cross-sectional study of 130 first-episode, drug-naïve MDD patients and 122 healthy controls. The data likely includes connectivity measures for rich club, feeder, and local connections in structural brain networks derived from diffusion imaging. The study investigates relationships between these network properties and childhood maltreatment scores.
A cross-sectional study from 2026 by Junhua Huang investigates the relationship between childhood maltreatment and rich club organization in structural brain networks. The dataset includes diffusion imaging data from 130 first-episode, drug-naïve MDD patients and 122 healthy controls. It contains connectivity measures (strength and density) for different connection classes, correlated with clinical scores.