mapbox/robosat
An end-to-end semantic segmentation pipeline for extracting features like buildings, roads, and water from aerial and satellite imagery.
Not currently ranked — collecting fresh signals.
star history
RoboSat is a Python-based machine learning pipeline for semantic segmentation on aerial and satellite imagery. It provides commands for training segmentation models, making predictions, and serving trained models. The pipeline extracts visually distinguishable features such as buildings, parking lots, roads, and cars from imagery and outputs structured GeoJSON polygons.
Frequently asked
- What is mapbox/robosat?
- An end-to-end semantic segmentation pipeline for extracting features like buildings, roads, and water from aerial and satellite imagery.
- Is robosat open source?
- Yes — mapbox/robosat is open source, released under the MIT license.
- What language is robosat written in?
- mapbox/robosat is primarily written in Python.
- How popular is robosat?
- mapbox/robosat has 2.1k stars on GitHub.
- Where can I find robosat?
- mapbox/robosat is on GitHub at https://github.com/mapbox/robosat.