With ArcGIS GeoAI tools, you can use deep learning pretrained models or train your own models to extract features from raw data, such as detecting trees, digitizing building footprints, or generating land-cover maps.
Install the deep learning libraries needed to run deep learning workflows in ArcGIS Pro, and learn to troubleshoot the most common issues.
Use a multi-purpose GeoAI model with a free-form language prompt to detect boats in Copenhagen imagery.
Use a GeoAI tool and a pretrained model to automate palm tree detection.
Use transfer learning to fine-tune a deep learning pretrained model in ArcGIS Pro and obtain enhanced results when extracting building footprints in a Seattle neighborhood.
Use a deep learning pretrained model to extract land cover from high-resolution drone imagery.
Extract building footprints from imagery using deep learning and apply raster functions to perform a landslide susceptibility analysis.
Use a deep learning pretrained model to extract water pixels from pre- and post- flood Sentinel-1 datasets, and perform change detection analysis to identify flooded areas in the St. Louis, Missouri region in 2019.
Perform automated damage assessment of homes after the devastating Woolsey fires.
Use deep learning to determine the extent of mangrove forests in Mumbai, India and how their footprints have changed over time.
Perform lidar point cloud classification using deep learning techniques to classify power lines.
Build and verify a model that can be used to automatically identify street signs with ArcGIS Survey123.
Use the Train Using AutoDL tool to train several deep learning models and pick the best performing one for a pixel-level land cover classification task.