GeoAI tutorial details building footprint extraction using U-Net, DINO, SAM, and Mask R-CNN

This tutorial details a GeoAI workflow for extracting building footprints from aerial imagery using a combination of deep learning models. It covers setting up the geospatial environment, training a U-Net model with a ResNet-34 encoder, and performing inference. The process also includes converting segmentation masks to polygons, evaluating performance metrics, and exploring zero-shot segmentation with Grounding DINO and SAM, as well as comparing results with a pre-trained Mask R-CNN model. AI

IMPACT Provides a practical guide for applying multiple AI models to extract building footprints from geospatial data.

RANK_REASON The item describes a tutorial and workflow for a specific application of AI, rather than a novel research paper or a product release. [lever_c_demoted from research: ic=1 ai=1.0]

Read on MarkTechPost →

• GeoAI
• Grounding DINO
• Mask R-CNN
• Microsoft Planetary Computer
• ResNet-34
• SAM
• U-Net

• product
• paper

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →