Archive: Training-data

Check out this repository for models using Facebook's Detectron2 to detect parking bays from satellite imagery and synthetic data.

AI-TOD is a dataset for tiny object detection in aerial images. The mean size of objects in AI-TOD is about 12.8 pixels, which is much smaller than in other datasets.

Check here for a dual stream U-Net architecture method for urban change detection using Sentinel-1 and Sentinel-2 data fusion.

Determine the location of a given street view image by matching it against a collection of geo-tagged satellite images. Read more about it here.

LEVIR-Ship is the first public tiny ship detection dataset specific to medium-resolution remote sensing images. Check it out here.

This is a single class dataset consisting of tiles of satellite imagery labeled with potential 'targets' to help with search and rescue.

This article analyses the Airbus oil storage dataset which contains over 13,500 annotated POL (petroleum, oil and lubricant) storage objects. This dataset is used to train an oil storage detector.

The AiTLAS toolbox includes state-of-the-art machine learning methods for exploratory and predictive analysis of satellite imagery as well as repository of AI-ready earth observation datasets.

Radiant Earth has released LandCoverNet, which will enable the creation of high-resolution and up-to-date maps for natural resource management.

This repository provides code for training and evaluating a convolutional neural network (CNN) to detect tree in urban environments with aerial imagery.