Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
16,405 datasets
mon_dataset_arabic_ocr provides a collection of images containing Arabic script for training and evaluating OCR models. The dataset's creator, size, and update date are unspecified. It originates from the Kaggle platform.
This dataset examines the relational and institutional foundations of seeking negative feedback from the perspective of inclusive leadership. It was authored by Lindomar Silva and is hosted by Harvard Dataverse. The specific number of rows, columns, and data structure are unknown.
Featuring video files of raw microscopy stacks for Hypsibius exemplaris embryos and hatchlings. The data was generated for visualization in an associated scientific paper by author Marc Mapalo.
605,866 anime-style illustrations were scraped from the Anime Pictures website as of June 2024. The collection is maintained by anime-pictures-archive and is updated via web crawling every three months. This dataset is intended for training image generation models and related research.
2026 data on elected officials in Calderdale, United Kingdom, provided by the Calderdale Metropolitan Borough Council. The dataset includes information on councillors, their associated wards, and political party affiliations.
DHBK_HB_EMB CNN GEI 045 V24.6 appears to be a dataset of Gait Energy Images (GEIs), which are silhouette-based motion templates used for human gait analysis. The cryptic title suggests it is version 24.6, likely containing 45 sequences or related to a specific model configuration, but the author, size, and creation date are unknown. It is hosted on Kaggle, indicating a focus on machine learning applications.
Labeled animal dataset designed for training YOLO object detection models. The dataset contains images annotated with bounding boxes for various animal species. Source, size, and creation details are unknown.
A dataset titled 'plant disease.v1i.yolov8' hosted on Kaggle. The platform tags suggest it contains images for object detection tasks related to plant disease identification. The dataset's specific contents, such as the number of images or annotation details, are not provided in the available metadata.
Weed detection 2.yolov8 is an image dataset formatted for training the YOLOv8 object detection model. The dataset contains annotated images of weeds, likely intended for agricultural computer vision tasks. Its origin and size are unspecified.
A Kaggle-hosted dataset containing checkpoint files for a HiFiGAN model. The dataset likely includes pre-trained model weights for generative audio synthesis tasks. Its specific size, creation date, and author are unknown from the provided metadata.
DCGAN dataset is a collection of images hosted on Kaggle, likely intended for training and evaluating Generative Adversarial Network models. The dataset's specific content, size, and origin are not detailed in the provided metadata. Its title suggests a focus on the DCGAN architecture, a popular model for image synthesis.
A Convolutional Neural Network model trained on Gait Energy Images, likely for human identification or analysis. The version is labeled as v24.6, indicating iterative development. The author and exact data source are unknown.
This dataset supports a book chapter analyzing barriers to the institutionalization of policy analysis. It includes raw data and appendices containing descriptives, full correlations, and interaction plots. The author is Prince Aian Villanueva.
Kaggle platform tags indicate this dataset relates to Generative Adversarial Networks (GANs) and synthetic data, likely for computer vision applications. The dataset's specific content, scale, and origin are not detailed in the provided metadata. Users must download the dataset to verify its exact composition and intended use.
ShiftGuard10 is a dataset for a computer vision challenge focused on robust image classification. The dataset likely contains images designed to test model performance against adversarial examples or distribution shifts. It is hosted on Kaggle, a platform for data science competitions.
Particle1904 released this repository in February 2026 to automate the selection, rescaling, and tagging of image-text pairs. It functions as a utility for preparing machine learning training data by standardizing visual and textual inputs.
A dataset of Formula 1 racing images, likely intended for computer vision tasks. It is published on Kaggle and tagged for object detection, suggesting it contains annotations for objects like cars, drivers, or track elements. The specific number of images, annotation format, and collection details are not provided in the available metadata.
ResNet50@WebFace600K is a dataset for face recognition tasks, likely containing images processed or used with the ResNet50 architecture. The dataset is published on Kaggle, a platform for data science and machine learning projects. Its specific size, origin, and update history are not detailed in the provided metadata.
Stroke Facial Image Dataset is a collection of facial images intended for medical analysis. The dataset is hosted on Kaggle and is tagged for tasks like binary classification and stroke detection. Specific details on the number of images, collection methodology, and licensing are not provided in the available metadata.
Eye Disease Balanced Image Dataset is a collection of medical images published on Kaggle. The dataset likely contains images related to various eye conditions. Metadata is minimal; actual content requires verification after download.