GeoAI – Inferencing & Post‑Processing in ArcGIS Pro (Part 4 of 4) скачать в хорошем качестве

GeoAI – Inferencing & Post‑Processing in ArcGIS Pro (Part 4 of 4)

In the final part of the GeoAI Workflow series, we take the trained tree classification models from Part 3 and move into inferencing, refinement, and QA/QC. You’ll learn how to: 00:00 Introduction 00:45 Create map & get ready 03:05 Explain different trained models & parameters 06:05 Run Classify Pixels Using Deep Learning to extract tree features 08:00 Compare SAMLoRA, DeepLabV3 & FasterRCNN outputs for tree canopy detection 12:15 Apply human‑in‑the‑loop edits and post‑processing to refine tree boundaries 15:05 Access post‑deep‑learning Python geoprocessing toolbox for advanced cleanup 16:30 Configure, run, and explore QA/QC tools to validate tree extraction accuracy 20:15 Closing and recap This stage completes the end‑to‑end GeoAI pipeline, delivering high‑quality tree extraction results ready for analysis and decision‑making. Resources shown in the video: Github GeoAI Smith ArcGIS Code Repo: https://github.com/ralouta/ArcGIS_Cod... GitHub Post‑Processing tool: https://github.com/ralouta/ArcGIS_Cod... Trees SAMLoRA dlpk: https://www.arcgis.com/home/item.html... Trees DeepLabV3 dlpk: https://www.arcgis.com/home/item.html... Trees FasterRCNN dlpk: https://www.arcgis.com/home/item.html...