Deep-learning

GeoTorch is a spatiotemporal deep learning framework on top of PyTorch and Apache Sedona. It enables spatiotemporal machine learning practitioners to implement deep learning models.

This project aims to provide composable Iterable-style and Map-style building blocks called DataPipes that work well out of the box with the PyTorch's DataLoader.

The Gym is a toolbox to segment imagery with a variety of a family of UNet models, facilitating fully reproducible label-to-model workflows.

This repository contains the code and configuration files to reproduce semantic segmentation results of Swin Transformer.

This article aims to demonstrate how to semantically segment aerial imagery using a U-Net model defined in TensorFlow.

The Replicable AI for Microplanning (RAMP) project is an open-source deep learning model that accurately digitizes buildings in low- and middle-income countries using satellite imagery.

Here are three deep learning models for predicting future satellite images from past ones using features such as precipitation and elevation maps.

This method aims to detect buildings and roads from pre-event satellite imagery while determining for each object instance whether it is affected by a recent flood event in post-event imagery.

Here is a huge list of resources for performing deep learning on satellite and aerial imagery. The resources are updated regularly and could benefit both the research and developer communities.

This website provides a comprehensive and interactive catalog of reference benchmark datasets. Check it out here.