Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
15,976 datasets
Uganda is the geographic scope of this dataset. It provides monthly forecasts of neonatal mortality for the period January to October 2024, generated using a Bayesian dynamic linear model. The dataset was authored by George Bamwebaze and published on figshare under a CC-BY-4.0 license.
Water temperature, salinity, nutrient concentrations, and radium isotope data were collected from the Yellow and East China Seas aboard the R/V Onnuri in February 2017. The dataset includes measurements of dissolved inorganic nitrogen, phosphorus, and 228Ra, captured using a Sea-Bird SBE 911-plus CTD and analyzed with an auto nutrient analyzer. Data is provided by the Korean Institute of Ocean Science and Technology and archived by NOAA NCEI.
Videoreason Training is a large-scale dataset comprising 471,575 samples across three subsets: perception (176,907 samples), simulation (105,818 samples), and embodied (188,850 samples). It was created by Zane-QIU and last updated on HuggingFace in April 2026. The dataset is designed for training models on video reasoning tasks spanning visual perception, 3D simulation, and robotics.
NE Atlantic observations from the 1989 North Atlantic Bloom Survey aboard the ATLANTIS II. The dataset contains three files with measurements of physical properties, nutrient chemistry, and biological pigments and organisms collected via bottle casts. It was submitted by Peter G. Brewer of the Woods Hole Oceanographic Institution as part of the Joint Global Ocean Flux Study.
Kaggle hosts the WLASL-300 dataset. It likely contains video data annotated with 300 keypoints for sign language recognition. The dataset's specific size, author, and last update date are unknown.
Kaggle hosts this dataset titled 'paddleocr-part1-output1'. The dataset likely contains output from an optical character recognition (OCR) pipeline, possibly from the PaddleOCR framework. Its specific contents, scale, and origin are not detailed in the provided metadata.
A dataset for training object detection models, likely containing images annotated for passenger detection. It is hosted on Kaggle and appears to be formatted for use with the YOLOv8 and YOLOv11 frameworks. The specific source, collection method, and temporal coverage are not detailed in the available metadata.
A dataset titled '0514 Organize Screwdriver 4 Clean' hosted on HuggingFace. It was uploaded by author juyoungggg and last updated on May 14, 2026. The title suggests the content may relate to organizing or cleaning screwdrivers, likely for a computer vision task.
4.1 million subject–text–video triples form this dataset for subject-driven video generation. Created by HiDream-ai, it was last updated in April 2026. It includes instance segmentation, face detection, quality scores, and timeline annotations.
ASID-1M is a large-scale audiovisual instruction dataset designed to support universal video understanding through fine-grained, controllable supervision. It addresses the limitations of traditional monolithic captions by providing attribute-structured and quality-verified data. The dataset aims to improve coverage of both visual and auditory elements within video content for more precise model training.
Public notices for procurement by the Imperial Household Agency of Japan, specifically for bids not covered by World Trade Organization agreements. The notices are published by the Accounting Section of the Secretariat of the Imperial Household Agency. The dataset's last recorded update was March 31, 2026.
Government procurement notices from Japan's Imperial Household Agency for tenders subject to World Trade Organization rules. The data is published by the Imperial Household Agency's Director-General of the Secretariat Accounting Division and was last updated on March 31, 2026. The notices cover general competitive bidding and selective competitive bidding processes.
Finite element numerical models and 3D printing codes related to the fabrication and oxygen transport within multi-compartment bioartificial pancreas devices. The dataset was contributed by author Hoesli, Corinne and was last updated on 2026-04-25.
21,500 synthetic images across 43 traffic sign classes produced by FraunhoferIOSB as a 'synthetic twin' to the German Traffic Sign Recognition Benchmark (GTSRB). Each class contains exactly 500 independent images to facilitate balanced training for traffic sign recognition tasks.
A 495-page book by Tony Smith analyzing U.S. foreign policy and the global struggle for democracy. The text includes 13 chapters covering historical periods from Wilson to Obama, plus extensive notes and a bibliography. The content is sourced from the paperswithcode platform.
Survey data measures the ability of Canadian businesses to take on more debt in the first quarter of 2026. It is broken down by industry classification, business employment size, type of business, activity, and majority ownership. The dataset is produced by Statistics Canada.
First quarter of 2026 survey data on business plans to apply for debt financing and the intended uses of that funding. The dataset is categorized by North American Industry Classification System (NAICS), business employment size, type of business, activity, and majority ownership. It is published by Statistics Canada and was last updated in February 2026.
Statistics Canada collected survey data on actions businesses took to address skills gaps in the first quarter of 2026. The dataset categorizes responses by North American Industry Classification System (NAICS), business employment size, type of business, business activity, and majority ownership. It was published by Statistics Canada on February 27, 2026.
Statistics Canada provides a dataset measuring the percentage of employees fully proficient in their current jobs for the first quarter of 2026. It is disaggregated by the North American Industry Classification System (NAICS), business employment size, type of business, business activity, and majority ownership. The data was last updated on February 27, 2026.
Statistics Canada provides a dataset on businesses or organizations that saw increased sales of Canadian products over the previous 12 months. The data is categorized by NAICS code, business size, type, activity, and ownership structure. It was last updated in the first quarter of 2026.