Esri Tutorials Spotlight: GeoAI
If you’re curious about how artificial intelligence and geographic information systems intersect, now is a great time to explore GeoAI in ArcGIS.
GeoAI in ArcGIS empowers users to extract, analyze, and predict from imagery, vector, text, and time‑series data with powerful built‑in tools and pretrained models.
In this blog post, we feature five Esri Tutorials designed to get you hands-on with object detection, deep learning, land cover extraction, and more.
Detect objects with Text SAM
In this tutorial, you’ll use Text SAM, a multipurpose GeoAI model, to automatically detect objects like boats in aerial imagery by prompting the model with plain text. This approach offers a flexible, prompt-based workflow that shows how natural language can guide AI-driven object detection directly in ArcGIS Pro.
Improve a Deep Learning Model with Transfer Learning
Pretrained deep learning models are powerful, but they perform best when your imagery is the same as the data they were trained on; differences in resolution, land cover, or building styles can lead to poor results. In this tutorial, you’ll learn how to take a pretrained model from ArcGIS Living Atlas and improve its performance using transfer learning — fine-tuning it with new samples to better detect building footprints in your specific area of interest.
Extract High-Resolution Land Cover with GeoAI
Understanding land cover is essential for environmental planning, ecosystem assessments, and change-detection analysis. This tutorial walks through how to extract high-resolution land cover information from imagery using GeoAI tools in ArcGIS Pro, giving you a pixel-level classification that can support everything from urban planning to natural resource monitoring.
Extract Informal Settlements with SAMLoRA
Urban planners and researchers often need to identify and map informal settlement areas quickly and accurately. This tutorial introduces SAMLoRA, a GeoAI model that blends foundation segmentation with low-rank adaptation to efficiently extract informal settlement extents from high-resolution imagery — helping you support social infrastructure and planning decisions.
Try GeoAI in ArcGIS
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. Explore this 13-step learning plan where you can explore how GeoAI is used in ArcGIS
------------------------------
GeoAI might sound intimidating at first, but these tutorials break everything down into guided steps so you can learn by doing. These tutorials make GeoAI approachable and beginner-friendly, while still teaching skills you can use for class projects, portfolio pieces, research, or internships.