Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
15,976 datasets
148 Sprague–Dawley male rats were used to investigate how NaV1.8 contributes to hyperalgesia induced by anxiety and remifentanil. Created by Jinxia Cai and updated in 2026, the data includes behavioral assessments and molecular measurements of mitochondrial dynamics.
12670 data points provide detailed individual survival records for water mites over a 14-week experimental period. The dataset serves as a foundational resource for studying water mites as experimental organisms in zoology. It is openly available under a CC-BY-4.0 license.
deadtrees.earth is a multi-resolution aerial image dataset for tree and mortality detection, hosted on Harvard Dataverse. The dataset was authored by Ayushi Sharma and was last updated on April 28, 2026. Its specific scale, geographic coverage, and collection method are not detailed in the provided metadata.
Records from NYC DOT Art detail temporary art installations on city property, including their location and duration. The dataset includes columns for installation and removal dates, artist, title, partner organization, borough, and site type. It is maintained by data.cityofnewyork.us and was last updated in March 2026.
yolo11s_dataset is a computer vision dataset hosted on Kaggle. The platform tags suggest it contains images intended for object detection tasks, likely to be used with YOLO (You Only Look Once) models. The dataset's specific content, size, and origin are not detailed in the available metadata.
IndoorCrowd is a multi-scene dataset designed for indoor human detection, instance segmentation, and multi-object tracking. It captures diverse challenges such as viewpoint variation, partial occlusion, and varying crowd density across four distinct campus locations. The dataset was created by author 'sebnae' and was last updated on 2026-04-02.
The U.S. Department of Housing and Urban Development (HUD) maintains a network of 10 regional offices and at least one field office in every state. This dataset describes the administrative hierarchy, with Regional Administrators overseeing regions and Field Office Directors managing local offices. The data was last updated on March 11, 2026, and is sourced from the agency's official content on Data.gov.
Fair Housing Initiatives Program (FHIP) grantees are private non-profit organizations funded to assist with housing discrimination issues. The dataset denotes their locations and pertinent information, sourced from the U.S. Department of Housing and Urban Development. It was last updated on March 11, 2026.
Turkish Number Plates.v2i.yolov8 is a dataset hosted on Kaggle. The title suggests it contains images of Turkish vehicle license plates, likely formatted for training the YOLOv8 object detection model. The dataset's specific size, creation date, and authorship details are not provided in the available metadata.
Car_colors_yolo is a dataset published on Kaggle. Its title suggests it contains images of cars, likely annotated for object detection tasks using the YOLO framework. The dataset's specific content, size, and origin are not detailed in the provided metadata.
Kaggle hosts a machine-learning-ready version of the Titanic passenger manifest. The dataset has been cleaned and includes engineered features for predictive modeling. Its author, organization, and last update date are unknown.
ResNet12_miniimagenet_pretrained(512) is a set of pretrained model weights hosted on Kaggle. The dataset likely contains parameters for a ResNet12 architecture trained on the MiniImageNet benchmark. Its specific content and size require verification after download.
Geoscience Australia Data provides a 2019 national gravity grid derived from 1.4 million ground gravity observations, 345,000 line km of airborne gravity, and 106,000 line km of gravity gradiometry data. The map presents a Hue-Saturation-Intensity image of De-trended Global Isostatic Residual Gravity data, with a linear color scale from -500 to +500 µm.s-2.
100 sample images from a larger 50,000-image synthetic dataset for 6-degree-of-freedom pose estimation. The dataset features the ABB dino robot model with 17 keypoints, rendered in Blender 5.1 EEVEE at 640×640 resolution. Annotations are provided in COCO keypoint format for the sample set.
OCR_license_plate likely contains images of vehicle license plates. The dataset is hosted on Kaggle. Specific details about the number of images, their source, or creation date are unknown.
A collection of medical images related to Polycystic Ovary Syndrome (PCOS). The dataset is hosted on Kaggle, but its specific size, collection dates, and authorship are not detailed in the available metadata. Columns and sample data are unknown, limiting immediate assessment of its structure and content.
A dataset for computer vision feature learning, likely associated with the YOLO (You Only Look Once) object detection framework. The dataset's specific content, size, and provenance are not detailed in the provided metadata. Published on Kaggle, it focuses on dual-morphology feature learning, suggesting a collection of images for training or evaluating feature extraction methods.
A collection of images likely intended for training object detection models, specifically for use with the YOLO framework. The dataset's title suggests a focus on eggs within an incubator environment. It is hosted on Kaggle, but specific details about its size, creation date, and author are unknown.
Chemical data on effluent and discharge was collected during the Beaufort Sea Monitoring Program. Measurements were taken from the Beaufort Sea over a 16-day period in August 1989 by Arthur D. Little, Inc. and submitted to the National Oceanic and Atmospheric Administration (NOAA). The processed data is stored in the NODC F144-Marine-Toxic-Substances file format.
Four benchmark datasets contain images of chemical Markush structures from patents and their corresponding CXSMILES string representations. The largest subset, 'uspto-mol-m-54k-new', includes 54,785 training samples. The datasets were created by docling-project and were last updated in March 2026.