Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
16,059 datasets
Predicted YOLO Set for Correction is a dataset hosted on Kaggle. The dataset likely contains images with object detection predictions generated by a YOLO model, intended for review and correction. Metadata is minimal; actual content requires verification after download.
100 fully synthetic, machine-readable invoice PDFs are provided, each paired with JSON and raw text files. The dataset is hosted on Kaggle, but the author, organization, and license details are unknown. The last update date is also unspecified.
A JSON dataset containing the text documents of Mahatma Gandhi. The dataset's author, organization, and last update date are unknown. Its size, row count, and file formats are unspecified.
Ash-free dry mass measurements for three size fractions of organic matter collected from benthic substrate samples. Data originate from up to 10 replicate samples per site across three English chalkstreams, collected on 5 to 7 occasions between October 2012 and October 2013. This dataset was created for quantifying food web mass and nutrient fluxes under the Wessex BESS project.
A public sample of 1,450 frames extracted from 10 video clips of professional soccer broadcasts. The dataset is curated for detecting and tracking soccer balls, specifically designed for tiny object detection challenges. It was created by infactory-ai and last updated on 2026-03-02.
Calderdale Metropolitan Borough Council provides a list of all business rateable properties, including relief indicators and rateable values. The dataset includes specific relief codes such as SBR, DCR, and MCR, and excludes properties where the ratepayer is an individual or partnership. It was last updated in March 2026.
A series of digital LiDAR LAS files managed by the State of Indiana Geographic Information Office and IOT Office of Technology. The data includes collections from 2011-2013 and an NRCS-funded collection from 2016-2020, organized into a tile grid covering the entire state. The files are stored uncompressed in AWS for cloud access and processing.
A series of digital orthophotography for the state of Indiana managed by the State Geographic Information Office and IOT Office of Technology. The imagery dates back to 2005 and is organized into a tile grid scheme covering the entire geography of the state. Each year's imagery is available in multiple formats including Cloud Optimized GeoTIFF (COG), original GeoTIFF, ECW, and MrSID.
The SPARC Datasets comprise a collection of scientific data focused on bridging the body and the brain. The data is produced by The SPARC Data and Resource Center and integrates molecular, imaging, and timeseries datatypes associated with neural connectivity and organ innervation.
A pretrained ResNet model likely intended for computer vision tasks in medical imaging. It was published on Kaggle, but its exact origin, training data size, and creation date are unspecified. The model's name suggests a connection to the VinBigData chest X-ray dataset.
RCNN_best_model is a dataset published on Kaggle. The title suggests it contains model weights or checkpoints for a Region-based Convolutional Neural Network (R-CNN). The dataset's specific content, size, and origin require verification after download.
FD-GAN is a dataset published on Kaggle. The dataset's content likely relates to Generative Adversarial Networks, a computer vision technique. Specific details regarding its size, creator, and temporal coverage are unavailable from the provided metadata.
Checkpoint-6CNN is a dataset published on Kaggle. The title suggests it contains saved model weights for a convolutional neural network, likely for image-related tasks. The dataset's author, organization, and specific contents are unknown.
Annotated images of smartphones formatted for object detection tasks. The dataset is structured according to the COCO (Common Objects in Context) annotation format, a widely used standard in computer vision. Its specific source, size, and creation details are not provided in the available metadata.
A multi-task inference benchmark for computer vision tasks. It likely contains annotated images of road defects under adverse weather conditions. The dataset provides semantic and instance masks for segmentation tasks.
Kaggle hosts a file containing the pre-trained weights for the ResNet50 convolutional neural network architecture. The file likely provides the learned parameters for image classification tasks. The original author, organization, and specific training data are unknown.
Kaggle hosts this dataset titled 'Data_Resnet152V2_global'. The title suggests it contains image data likely intended for use with the ResNet-152 V2 convolutional neural network architecture. The dataset's author, organization, size, and specific content are not provided in the metadata.
Red--White-wine-Dataset contains measurements of wine properties related to perceived quality. The data was obtained from a third-party industry organization for analysis by Digitas Advanced Analytics on behalf of Unicorn Winery. The dataset is shared under a CC0-1.0 license.
Department of Finance reports from 2012-2018 used to value NYC condominiums as if they were rental apartment buildings. The data includes property location, building classification, total units, year built, square footage, and estimated income and expenses. This valuation method is required by New York State law.
Latin America is the focus of this policy brief, which synthesizes data on economic crimes from 2020 to 2023. It likely contains tabular data on categories including corruption, money laundering, fraud, tax evasion, and organized crime. The dataset is intended for policy analysis and research on illicit financial activities in the region.