Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
16,039 datasets
A National Park Service Geologic Resources Inventory product, this digital map integrates GIS data layers and tables from U.S. Geological Survey and Idaho Geological Survey source maps. It is available in multiple GIS formats, including an ESRI file geodatabase, an OGC geopackage, and a KMZ file for Google Earth. The dataset includes ancillary PDF documents with geologic unit descriptions, metadata, and readme instructions for proper use.
Genome-wide identification of the Delay Of Germination 1 (DOG1) gene family in Brassica napus (rapeseed). The dataset likely contains gene expression patterns and transcriptome sequencing data related to manganese stress response. Its cross-platform presence on figshare suggests it is a supplementary resource for the associated research publication.
Alice is an AI-powered dataset management toolkit designed for YOLO (You Only Look Once) computer vision workflows. The toolkit, authored by simoncirstoiu, was last updated on 2026-04-26. Its description suggests capabilities for analyzing, learning from, ingesting, curating, and exporting datasets.
Manly 2020 3U OCR Outputs is a dataset published on Kaggle. The title suggests it contains text outputs from an Optical Character Recognition process. The specific source, size, and content require verification after download.
A survey on robust image segmentation under various degradation conditions, authored by CharlesChen2023. The dataset page was last updated on March 7, 2026. The description frames segmentation as a core task for visual understanding, Physical AI, and World Models.
Yupp AI collected organic user preferences from side-by-side comparisons of AI-generated SVG outputs. The dataset contains user chat logs with model responses and preference comparisons, providing a direct evaluation of model reasoning and coding capabilities. This release represents only a small fraction of the SVG data, as noted in the description.
Synthetic video frames for training object detection models to identify robbery threats at ATMs. The dataset likely contains images annotated with bounding boxes for guns and knives. It is hosted on Kaggle, but details on the creator, size, and update date are unknown.
A cleaned image dataset for classifying tomato leaf diseases. The dataset contains images across 10 distinct disease categories. It was sourced from Kaggle, but details on the author, collection method, and specific size are not provided.
EPP-YOLO26-3 is a dataset hosted on Kaggle. The title suggests a focus on computer vision, likely for object detection tasks. No further descriptive metadata, sample data, or column information is available.
Kaggle hosts a dataset titled 'Panoptic Segment'. The dataset likely contains images annotated for panoptic segmentation, a computer vision task that unifies semantic and instance segmentation. Its specific contents, size, and origin are not detailed in the provided metadata.
A dataset titled 'custom_resnet50_trashnet' is hosted on Kaggle. Its content likely contains images for classification tasks, inferred from the model name 'ResNet50' and the 'TrashNet' reference. The dataset's specific size, origin, and creation date are unknown.
Kaggle hosts this dataset focused on Generative Adversarial Network training. The dataset likely contains images or image-related features used for training GAN models, as inferred from the title. Specific details on volume, authorship, and creation date are unavailable.
A dataset named FD-GAN, likely related to Generative Adversarial Networks for computer vision tasks. It is published on the Kaggle platform. The specific content, size, and creation details are not provided in the available metadata.
FRFInal_FDGAN is a dataset published on Kaggle. The title suggests a connection to Generative Adversarial Networks (GANs), likely for image generation or manipulation. No further metadata on size, origin, or specific content is available.
A dataset titled 'conditional-gan2d-normalized' is hosted on Kaggle. The dataset likely contains normalized 2D image data generated for training or evaluating conditional Generative Adversarial Networks. No further metadata on size, origin, or specific content is available.
Kaggle hosts a dataset titled 'conditional-gan2d-unnormalized'. The dataset likely contains image data intended for training conditional Generative Adversarial Networks. Its author, organization, and update history are unknown.
PALSAR-2 ScanSAR data was collected by JAXA in response to Tropical Cyclone Mocha forming on 11 May 2023 and making landfall near Sittwe, Myanmar, with winds up to 175 km/h. The dataset contains orthorectified, 25-meter resolution normalized backscatter data from ALOS-2's broad area observation mode, covering a width of 350 km. Data is provided as 16-bit digital numbers convertible to gamma naught values in decibels.
Data from cruise NBP01-1 includes measurements of biogenic silicon, organic carbon, organic nitrogen, and C13/N15 isotopes from kasten and piston cores. Radiocarbon dates are provided for bulk organic matter, foraminifera, and macrophytes/macrofossils collected from these cores. The dataset originates from the SCIOPS organization and is hosted on NASA EarthData.
Historical oceanographic profile data for the Adriatic Sea contains measurements of temperature, salinity, chlorophyll, nutrients, and oxygen from the surface to depths exceeding 1000 meters. The dataset was collected starting in 1911 by Italian, U.S., and former Yugoslavian sources and is provided by IMGA-CNR (Institute for the Study of Geophysical and Environmental Methodologies, Italian National Research Council).
NASA Earthdata hosts a dataset on seafloor sediment and biota from Explorers Cove and McMurdo Sound, Antarctica. The data includes measurements of sediment granulometry, organic carbon, chlorophyll-a, and seafloor images collected during the 1999/2000 and 2001/2002 field seasons. It was produced by the organization SCIOPS for characterizing the locality for long-term ecosystem monitoring.