Building Footprint Extraction

The Building Footprints USA deep learning model is developed to extract building footprints. While it's designed to work in continental US, the model is seen to perform fairly well in other parts of the world. The resultant footprints can be used for a variety of purposes, including base map preparation, humanitarian aid, disaster management, and transportation planning. Applications range from taxation, urban planning, public works, visualization, simulation to a wide range of suitability analysis applications.

"Machine Intelligence is the last Invention Humanity will ever have to make" - SupplyChainToday.com"

This model is expected to perform best with high resolution aerial/satellite imagery (10 - 40 cm resolution)

The model can be downloaded by anyone who has an ArcGIS Online subscription. The following licenses are required to consume or run the model.

• Desktop workflows - Image Analyst Extension for ArcGIS Pro.

• Enterprise workflows - ArcGIS Image Server license with raster analytics configured. 

• Post processing (to regularize the resultant footprints) - 3D Analyst Extension 

The model can be used with the  Detect Objects Using Deep Learning  tool. A detailed tutorial on using the model in ArcGIS can be found  here  and a video for the same can be found  here .

Postprocessing (Optional): The tool  Regularize Building Footprint  can be used to improve the results.