# Microsoft Building Footprints

Worldwide building footprints derived from satellite imagery

Repository: https://github.com/microsoft/GlobalMLBuildingFootprints
Canonical: https://ross.abutalabs.com/products/microsoft-building-footprints
Language: Python
License: NOASSERTION
License Family: other
Last push: 2026-08-14T14:32:42+00:00

## Health v2 (maintenance only)
Score: 76/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 35, longevity 100
- inputs: {"age_days": 1594, "days_push": 19, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1940, forks 276 (observed 2026-08-28T04:05:57.279175+00:00)

## What it is
Open datasets of computer-generated building footprint polygons for the United States (and globally) derived by Microsoft using deep learning on satellite imagery. The US dataset contains over 129 million footprints in GeoJSON format, released under the ODbL license to support the OpenStreetMap ecosystem.

## Use cases
- download building footprints for all US states as GeoJSON
- import computer-generated building outlines into OpenStreetMap
- analyze building density and coverage for urban planning research
- use building polygons as training or evaluation data for computer vision models
- enrich GIS applications with nationwide building geometry data
- support humanitarian mapping with AI-assisted tasking

## When to choose
- you need large-scale, freely licensed building footprint polygons for the US or globally
- you want GeoJSON data ready for GIS tools or OSM workflows
- you need building geometry with associated capture-date metadata

## When to avoid
- you need hand-verified or survey-grade building boundaries
- you require frequently refreshed, near-real-time footprint data
- you need data under a permissive license without ODbL share-alike obligations

## Facets
- artifact type: dataset
- maturity: active
- function: machine-learning, computer-vision, geospatial, data-science
- domain: computer-vision, artificial-intelligence
- platform: cross-platform
- tags: building-footprints, satellite-imagery, geo, open-data, openstreetmap, semantic-segmentation, microsoft, maps, geospatial

## Member repositories
- microsoft/GlobalMLBuildingFootprints (main) score 76
- microsoft/USBuildingFootprints (mirror) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:57.279175+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:43:48.315877+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/GlobalMLBuildingFootprints (fetched 2026-08-28T04:05:57.279175+00:00, sha 10a1311c9daa)
- Data as of 2026-08-30T08:39:29.467469+00:00.