Home › AI & machine learning › GeoOSAM
GeoOSAM QGIS 4 compatibility
Not QGIS 4 ready Active
Not yet — the latest version declares max QGIS version 3.99.0, so it is not offered to QGIS 4 users.
AI-powered segmentation for QGIS — SAM2.1 & SAM3 with text prompts, similar object detection, vector ROI, and professional GIS workflow features
About
GeoOSAM brings Meta's SAM 2.1 and SAM3 to QGIS with a streamlined 4-tab UI (Segment, Detect, Results, Filters). Point-click or draw a bounding box on any raster to instantly segment objects. SAM3 users unlock the Detect tab: run text-prompt auto-segmentation ("find all vehicles"), click a reference object to find similar ones across the scene, or use any existing polygon as an exemplar via Similar from Selection. Pro users can run these modes over entire rasters with automatic tiling and restrict processing to a vector polygon ROI. Includes 12 pre-defined classes with color coding, custom class creation, real-world size and shape filters (GSD-based), multi-point segmentation (Shift/Ctrl+click), multi-format export (GeoPackage, Shapefile, GeoJSON), undo support, and persistent output folder. Optimized for NVIDIA GPU, Apple Silicon, and multi-core CPU with automatic hardware detection and model selection.
| Author | Ofer Butbega |
|---|---|
| Downloads | 17,271 |
| Latest version | 1.4.3 |
| Last updated | 2026-06-12 (125 days ago) |
| Min QGIS | 3.16.0 |
| Max QGIS | 3.99.0 |
| QGIS 4 ready | No |
| Maintenance | Active |
| Category | AI & machine learning |
| Tags | ai, remote sensing, gis, satellite imagery, python, segmentation, urban planning, machine learning, sam2.1, sam, pytorch, cuda, performance optimization, geospatial analysis, sam2, apple silicon, environmental monitoring, raster analysis, gpu acceleration, computer vision, object detection, deep learning, intelligent model selection, semantic segmentation, ultralytics, sam2.1_b |
Homepage · Repository · Issue tracker · Official page
Alternatives in AI & machine learning
Raster classification in QGIS with 11 machine learning algorithms, Optuna hyperparameter search, SHAP explanations, class imbalance handling
- ArcGeek CalculatorQGIS 4 ready
ArcGeek Calculator is a QGIS plugin that provides a comprehensive set of tools for coordinate calculations, conversions, spatial operations,
- Bunting Labs AI VectorizerQGIS 4 ready
Autocomplete for vectorizing and georeferencing raster maps.
Related plugins
- Geo SAMQGIS 4 ready
Segment Landform with Segment-Anything AI Model (SAM)
- AI Segmentation by TerraLabQGIS 4 ready
AI segmentation in QGIS 3.22 to 4. Draw a zone, name an object and get every building footprint, tree or solar panel as vector polygons, or
- TreeEyedNot QGIS 4 ready
TreeEyed is a QGIS plugin for tree monitoring using AI.
- AI Edit by TerraLabQGIS 4 ready
Generative AI image editing for QGIS 3.22 to 4. Show a redevelopment, a planted street, a flood or an orthophoto redrawn as a clean site pla
- GeoSeg StudioQGIS 4 ready
Full-pipeline deep learning semantic segmentation for geospatial raster data.
- AITracer by LADQGIS 4 ready
Click any feature on a georeferenced raster and SAM2 segments it instantly. Refine with positive/negative prompts, then accept to create a v
Maintenance: Active = updated within 1 year, Quiet = 1–3 years, Likely abandoned = over 3 years.