Min
Max
Min
Max
Min
Max
Min
Max
Top Google Datasets and Models
The datasets below can be used to train fine-tuned models for google detection. You can explore each dataset in your browser using Roboflow and export the dataset into one of many formats.
At the bottom of this page, we have guides on how to train a model using the google datasets below.
252 images1 model
Asus ROG Phone 7Google Pixel 8 ProHonor Magic5 ProHuawei P60 ProInfinix Zero UltraLenovo Legion Y90Motorola Edge 40Nokia G60OnePlus 11Oppo Find X6 ProPhonePoco F5 ProRealme GT Neo 5Redmi Note 12 Pro+Samsung Galaxy S23 UltraSony Xperia 1 VTecno Phantom V FoldVivo X90 ProXiaomi 13 ProZTE Axon 40 Ultra
105 images
fa fa-homefa-brands fa-googlefa-brands fa-youtubefa-sharp fa-solid fa-envelopes-bulkfa-solid fa-headphonesfa-solid fa-listfa-solid fa-microphone-slashfa-solid fa-trophyfa-solid fa-user-graduatefab fa-facebook-messengerfab fa-facebook-squarefab fa-instagramfab fa-linkedin-infab fa-telegramfab fa-twitterfab fa-whatsappfas fa-arrow-downfas fa-bellfas fa-bookfas fa-briefcase
800 images
(x1,y1),(x2,y2),aGong Cheng, Junwei Han, Peicheng Zhou, Lei Guo. Multi-class geospatial object detection and geographic image classification based on collection of part detectors. ISPRS Journal of Photogrammetry and Remote Sensing, 98: 119-132, 2014.Gong Cheng, Junwei Han. A survey on object detection in optical remote sensing images. ISPRS Journal of Photogrammetry and Remote Sensing, 117: 11-28, 2016.Please cite the following relevant papers when publishing results that use this dataset fully or partly:The folder "ground truth" contains 650 separate text files and each one corresponds to an image in "positive image set" folder. Each line of those text files defines a ground truth bounding box in the following format:These images were cropped from Google Earth and Vaihingen data set and then manually annotated by experts. The Vaihingen data was provided by the German Society for Photogrammetry, Remote Sensing and Geoinformation (DGPF): http://www.ifp.uni-stuttgart.de/dgpf/DKEPAllg.html.This dataset contains totally 800 VHR remote sensing images, where the folder "negative image set" includes 150 images that do not contain any targets of the given object classes and the folder "positive image set" includes 650 images with each image containing at least one target to be detected.This is a 10-class geospatial object detection dataset used for research purposes only.These ten classes of objects are airplane, ship, storage tank, baseball diamond, tennis court, basketball court, ground track field, harbor, bridge, and vehicle.This very-high-resolution (VHR) remote sensing image dataset was constructed by Dr. Gong Cheng et al. from Northwestern Polytechnical University (NWPU).where (x1,y1) denotes the top-left coordinate of the bounding box, (x2,y2) denotes the right-bottom coordinate of the bounding box, and a is the object class (1-airplane, 2-ship, 3-storage tank, 4-baseball diamond, 5-tennis court, 6-basketball court, 7-ground track field, 8-harbor, 9-bridge, 10-vehicle).
346 images
