mapbox/robosat
Semantic segmentation on aerial and satellite imagery. Extracts features such as: buildings, parking lots, roads, water, clouds
Based on the README, here's a technical summary: Implements a complete end-to-end pipeline using fully convolutional neural networks for pixel-level segmentation, with specialized tools for data preparation (downloading imagery from Mapbox APIs, extracting OSM geometries), model training on GPU/CPU, and post-processing that transforms segmentation outputs into cleaned GeoJSON features while handling Slippy Map tile boundaries. Works with standardized tile formats (256x256 pixels) to abstract geo-referenced imagery, and provides extensibility for custom data sources and feature types beyond the built-in extractors.
2,052 stars. No commits in the last 6 months.
Stars
2,052
Forks
388
Language
Python
License
MIT
Category
Last pushed
Aug 27, 2020
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/mapbox/robosat"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
ucam-eo/tessera
[CVPR26] TESSERA is a foundation model that can process time-series satellite imagery for...
arplaboratory/satellite-thermal-geo-localization
[IROS 2023] Official repository for "Long-range UAV Thermal Geo-localization with Satellite Imagery"
venkanna37/Label-Pixels
Label-Pixels is the tool for semantic segmentation of remote sensing images using Fully...
HakaiInstitute/habitat-mapper
Segmentation Tools for Remotely Sensed RPAS Imagery
WangLibo1995/BuildFormer
Building Extraction from remote sensing image using Vision Transformer, IEEE Transactions on...