GeoAI Course Resources

🌍 Essential Background and Foundational Resources

📘 Books and Key Overviews

Handbook of Geospatial Artificial Intelligence (Gao, Hu, Li, 2023) – Comprehensive reference with 30+ topics, code examples, datasets.
Deep Learning for the Earth Sciences (Camps-Valls, Tuia, Zhu, 2021) – Earth data meets deep learning; ideal for understanding EO models.
GeoAI and Human Geography (Xiao Huang, 2025) – Connects GeoAI with environmental and human systems.

📄 Review Articles

Boutayeb et al., 2024 – A comprehensive survey of GeoAI methods and applications.

Week-by-Week Resources

Week 1: Introduction to Geospatial Foundation Models

🌐 Development Seed Blog (2024): Using Foundation Models for Earth Observation
🚀 NASA/IBM Release: Prithvi HLS Foundation Model
☁️ AWS ML Blog (2025): Deploying GeoFMs on AWS

Week 2: Working with Geospatial Data

📚 Book: Learning Geospatial Analysis with Python (4th ed., 2023)
🧰 TorchGeo Docs: https://pytorch.org/geo
🌍 OpenGeoAI Toolkit: https://github.com/opengeos/geoai

Week 3: Loading & Using Foundation Models

🤗 Hugging Face Prithvi Models: https://huggingface.co/ibm-nasa-geospatial
📓 Demo Notebooks: Available on Hugging Face model cards.
🧪 AWS GeoFM Deployment Examples: Code and pipelines via GitHub.
🧠 IEEE GRSS Blog: Self-Supervised Geospatial Backbones

Week 5: Model Adaptation & Evaluation

🔌 Adapters for GeoFM: Explained via Development Seed blog.
📊 Evaluation Metrics: IoU, mAP, F1 from Boutayeb et al. (2024) + SpaceNet.
📁 Radiant Earth MLHub: https://mlhub.earth – Public geospatial ML datasets and benchmarks.

Interactive Sessions

1️⃣ PyTorch & TorchGeo Fundamentals

TorchGeo docs & sample loaders (CRS, tile sampling, multispectral).

2️⃣ Loading & Using Foundation Models

Hugging Face GeoFM models (Prithvi, Clay) and notebook demos.

4️⃣ Model Adaptation & Evaluation

Fine-tuning using TerraTorch and evaluation on public datasets.

5️⃣ Scalable Analysis & Deployment

📚 Geospatial Data Analytics on AWS (2023)
☁️ AWS SageMaker + GeoFM repo (scalable inference)

📦 General Tools & Repos

🔧 OpenGeoAI: https://github.com/opengeos/geoai
🛰️ IBM/NASA TerraTorch: https://github.com/ibm/terrapytorch
📍 Radiant MLHub Datasets: https://mlhub.earth
🧪 SpaceNet Challenges: https://spacenet.ai/challenges/

🧭 Deployment and Project Resources

🔧 **Flask/Streamlit for D

--- title: "GeoAI Course Resources" description: "Mapped books, tutorials, and online resources for a graduate research seminar on GeoAI" author: "Kelly Caylor" date: "2025-08-06" format: html --- ## 🌍 Essential Background and Foundational Resources ### 📘 Books and Key Overviews - **Handbook of Geospatial Artificial Intelligence** (Gao, Hu, Li, 2023) – Comprehensive reference with 30+ topics, code examples, datasets. - **Deep Learning for the Earth Sciences** (Camps-Valls, Tuia, Zhu, 2021) – Earth data meets deep learning; ideal for understanding EO models. - **GeoAI and Human Geography** (Xiao Huang, 2025) – Connects GeoAI with environmental and human systems. ### 📄 Review Articles - **Boutayeb et al., 2024** – A comprehensive survey of GeoAI methods and applications. --- ## Week-by-Week Resources ### Week 1: Introduction to Geospatial Foundation Models - 🌐 **Development Seed Blog (2024)**: [Using Foundation Models for Earth Observation](https://developmentseed.org/blog/2024/11/15/geoai-foundation-models/) - 🚀 **NASA/IBM Release**: [Prithvi HLS Foundation Model](https://www.earthdata.nasa.gov/news/geoai-foundation-model) - ☁️ **AWS ML Blog (2025)**: [Deploying GeoFMs on AWS](https://aws.amazon.com/blogs/machine-learning/revolutionizing-earth-observation-with-foundation-models/) ### Week 2: Working with Geospatial Data - 📚 **Book**: *Learning Geospatial Analysis with Python (4th ed., 2023)* - 🧰 **TorchGeo Docs**: [https://pytorch.org/geo](https://pytorch.org/geo) - 🌍 **OpenGeoAI Toolkit**: [https://github.com/opengeos/geoai](https://github.com/opengeos/geoai) ### Week 3: Loading & Using Foundation Models - 🤗 **Hugging Face Prithvi Models**: [https://huggingface.co/ibm-nasa-geospatial](https://huggingface.co/ibm-nasa-geospatial) - 📓 **Demo Notebooks**: Available on Hugging Face model cards. - 🧪 **AWS GeoFM Deployment Examples**: Code and pipelines via GitHub. - 🧠 **IEEE GRSS Blog**: [Self-Supervised Geospatial Backbones](https://grss-ieee.org/blog/2023/04/self-supervised-backbones/) ### Week 4: Multi-modal & Generative Models - 🛰️ **Sensor Fusion via TorchGeo**: Stack Sentinel-1/2 & others. - 🌐 **Presto, Clay, DOFA Models**: Explained in DevSeed & AWS blogs. - 🧪 **DiffusionSat Example**: [Medium Tutorial on Generating EO Images](https://medium.com/@karray/diffusion-models-for-remote-sensing-part-5-abc123) ### Week 5: Model Adaptation & Evaluation - 🔌 **Adapters for GeoFM**: Explained via Development Seed blog. - 📊 **Evaluation Metrics**: IoU, mAP, F1 from Boutayeb et al. (2024) + SpaceNet. - 📁 **Radiant Earth MLHub**: [https://mlhub.earth](https://mlhub.earth) – Public geospatial ML datasets and benchmarks. --- ## Interactive Sessions ### 1️⃣ PyTorch & TorchGeo Fundamentals - TorchGeo docs & sample loaders (CRS, tile sampling, multispectral). ### 2️⃣ Loading & Using Foundation Models - Hugging Face GeoFM models (Prithvi, Clay) and notebook demos. ### 3️⃣ Multi-modal & Sensor Fusion - DOFA & Clay examples; stacking Sentinel data with TorchGeo. ### 4️⃣ Model Adaptation & Evaluation - Fine-tuning using TerraTorch and evaluation on public datasets. ### 5️⃣ Scalable Analysis & Deployment - 📚 *Geospatial Data Analytics on AWS* (2023) - ☁️ AWS SageMaker + GeoFM repo (scalable inference) --- ## 📦 General Tools & Repos - 🔧 **OpenGeoAI**: [https://github.com/opengeos/geoai](https://github.com/opengeos/geoai) - 🛰️ **IBM/NASA TerraTorch**: [https://github.com/ibm/terrapytorch](https://github.com/ibm/terrapytorch) - 📍 **Radiant MLHub Datasets**: [https://mlhub.earth](https://mlhub.earth) - 🧪 **SpaceNet Challenges**: [https://spacenet.ai/challenges/](https://spacenet.ai/challenges/) --- ## 🧭 Deployment and Project Resources - 🔧 **Flask/Streamlit for D