opengeos/geoai

GeoAI: Artificial Intelligence for Geospatial Data

What it solves

GeoAI simplifies the process of applying artificial intelligence to geospatial data. It addresses the fragmentation of tools by providing a unified framework that abstracts the complex machine learning workflows required for analyzing satellite imagery, aerial photographs, and vector data, making these techniques accessible to researchers without deep ML expertise.

How it works

GeoAI integrates several AI frameworks (PyTorch, Transformers, PyTorch Segmentation Models) and specialized geospatial libraries. It provides high-level APIs for the entire AI lifecycle: searching and downloading remote sensing imagery, preparing datasets with image chips and labels, training models for classification, detection, and segmentation, and running inference pipelines. It also includes a curated catalog of remote sensing foundation models and integrates with Leafmap and MapLibre for visualization.

Who it’s for

  • Geospatial researchers who need AI workflows without deep machine learning expertise.
  • AI practitioners seeking streamlined geospatial preprocessing and domain-specific datasets.
  • Educators looking for reproducible, teaching-ready workflows for geospatial AI.

Highlights

  • Unified AI Workflow: Covers everything from data acquisition to inference and visualization.
  • Curation of Foundation Models: Provides a catalog of 20 remote sensing foundation models with metadata and loading capabilities.
  • QGIS Integration: A dedicated plugin allows users to run AI-powered workflows directly within the QGIS desktop environment without writing code.
  • Broad Data Support: Supports multiple formats including GeoTIFF, GeoJSON, Shapefile, and GeoPackage.
  • GPU Acceleration: Includes automatic device management for GPU acceleration.

Related

  • Project
    microsoft/farmvibes-ai

    A multi-modal geospatial machine learning framework for agriculture and sustainability that fuses satellite imagery, weather data, and elevation maps to generate robust agricultural insights.

  • Project
    NVIDIA/earth2studio

    A Python-based AI inference pipeline toolkit for weather and climate science that provides a unified API to integrate various AI models and data sources.

  • Project
    Esri/deep-learning-frameworks

    A collection of deep learning and machine learning library installers that integrate PyTorch, transformers, and other AI frameworks into the ArcGIS ecosystem for geospatial analysis.

  • Project
    opengeos/GeoAgent

    A shared AI agent layer that enables geospatial Python packages and QGIS plugins to expose their tools to LLMs through a consistent interface.

  • Project
    jd-opensource/JoyAI-Image

    A unified multimodal foundation model that integrates image understanding, text-to-image generation, and instruction-guided image editing with a focus on spatial intelligence.