Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
16,012 datasets
DeepSeek-OCR-Code is a dataset hosted on Kaggle, likely containing images of code snippets paired with their corresponding text. The dataset's title suggests a focus on applying optical character recognition techniques specifically to programming code. Its specific scale, origin, and update history are not detailed in the available metadata.
Hindi OCR Lines is a dataset for optical character recognition tasks, likely containing images of text lines in the Hindi script. It is hosted on Kaggle, but the author, organization, and specific collection details are unknown. The dataset's size, format, and exact contents require verification after download.
A dataset of synthetic images likely generated by Generative Adversarial Networks (GANs). It originates from the Kaggle platform, but the author, organization, and specific creation date are unknown. The dataset's connection to 'CICIoV2024' suggests it may be related to a computer vision challenge or event from 2024.
FasterRCNN_task1 is a dataset hosted on Kaggle, likely for training or benchmarking object detection models using the Faster R-CNN architecture. The dataset's author, organization, size, and specific contents are not detailed in the provided metadata. Its last update date is unknown.
An image dataset hosted on Kaggle, likely intended for computer vision tasks. The dataset's specific content, size, and collection details are not provided in the metadata. Its origin, creation date, and update history are unknown.
MaskRCNN_Phase1_weights is a set of pre-trained weights for a Mask R-CNN model, a popular architecture for instance segmentation. The dataset is hosted on Kaggle, a platform for data science and machine learning projects. The specific source, training data, and performance metrics for these weights are not detailed in the available metadata.
SVHN is a real-world image dataset for developing machine learning and object recognition algorithms. It is derived from Google Street View imagery and is a popular benchmark in computer vision. The dataset is published on Kaggle.
PadangFood v2 is an image classification dataset hosted on Kaggle. The title suggests it contains images of food, likely from the Padang culinary tradition of Indonesia. The dataset's size, author, and specific contents are not detailed in the available metadata.
ICG_spacenet_sandox_yolo is a dataset published on Kaggle. The title suggests it contains annotations formatted for the YOLO object detection framework, likely derived from SpaceNet satellite imagery. The dataset's specific content, scale, and origin require verification after download.
A corpus of computer science paper abstracts sourced from the arXiv API. The abstracts are organized into four thematic categories, including Algorithms and NLP/AI. The dataset was collected for an NLP Lab at IIT Jammu during the 2025-2026 academic year.
Local Health Districts Status Maps from the State of Connecticut provides administrative and contact details for local health departments. The dataset includes department names, planning organizations, director names, titles, degrees, emails, statuses, and phone/fax numbers. It was last updated on 2026-03-22 03:00:32.008276.
The Arctic Ocean surface meteorological data from Western Arctic ice drifting stations, including AIDJEX, ARLIS I, ARLIS II, Ice Station Alpha, Ice Station Charlie, T-3, and the ships Maud and Fram. These data were collected and organized by NCDC and NSIDC, leading to the production of a CD-ROM. Several different versions of the data are available, extending from the original key-entered data to variable subsets and QC versions developed in uniform format for all stations.
1992 data from a study analyzing the chemical composition of peat deposits from Macquarie Island. The dataset contains measurements of mono- and dicarboxylic acids, categorized by saturation state, from samples subjected to hydrothermal pyrolysis at four temperatures. It was produced by researchers S. A. Pickering and B.D. Batts based on samples collected at Green Gorge.
VCBench contains 4,574 clipped video segments totaling approximately 80 GB, developed by buaaplay and last updated in March 2026. The collection is organized into 8 subcategories specifically designed to evaluate spatial-temporal state maintenance in video understanding models. It focuses on two primary task types: object counting and event counting.
Fifty CTD casts and nutrient samples were collected in the Prydz Bay region during the Antarctic Division BIOMASS Experiment III (ADBEX III) cruise of the Nella Dan from September to December 1985. The dataset contains measurements for pressure, temperature, salinity, and nutrients like nitrate and phosphate. It was contributed by the Australian Antarctic Data Centre (AU_AADC).
An investigation of desert varnish distribution in Antarctica's Wright Valley was conducted by SCIOPS. Traverses on foot revealed the coating is found almost exclusively in high altitude areas and is dominantly associated with dolerite rock. The data originates from a study published in February 1973.
1989 data from the Okinawa Trough details the distribution and abundance of particulate DNA and carbon around the Izena black smoker hydrothermal vent. The dataset, produced by SCIOPS, also reveals features of culturable microorganisms in the area. Spatial coverage is limited to a specific bounding box defined by latitude and longitude coordinates.
Annual monitoring data tracks the status and trends of coral reef populations in the Florida Keys National Marine Sanctuary. The dataset includes species counts, percent cover of corals and other organisms, and coral disease incidence from 40 sampled sites. Data was collected by the Coral Reef Monitoring Project, with the last recorded update in 1999.
Samples collected at O'Gorman Rocks and Ellis Fjord near Davis station from December 1997 to March 1998 provide data on zooplankton abundance, chlorophyll a concentration, and particulate organic carbon. The dataset resulted from projects ASAC 963 and ASAC 2229, managed by the Australian Antarctic Data Centre. Field and laboratory work included depth-stratified sampling, sediment trap collection, and grazing experiments.
A historical record of all thunderstorms reported in weather observations at Cape Kennedy Air Force Station, Florida. The dataset spans 16 years from 1957 to 1972 and was compiled by NOAA's National Centers for Environmental Information. It documents storm events with detailed meteorological elements.