No Clocks Vault

List - Building Footprint Datasets

2 min read

List - Building Footprint Datasets

Overview

Curated collection of open building footprint datasets from major providers. These datasets are essential for property analysis, urban planning, and geospatial applications.

Major Providers

Microsoft Building Footprints

AI-extracted building polygons covering the United States.

AspectDetails
CoverageUnited States
Buildings129+ million
FormatGeoJSON
LicenseODbL (Open Database License)
GitHubmicrosoft/USBuildingFootprints
Globalmicrosoft/GlobalMLBuildingFootprints

Features:

  • High-quality ML-derived footprints
  • State-by-state downloads
  • Regular updates
  • Global dataset also available (1.2B+ buildings)

Google Open Buildings

Global building footprints derived from satellite imagery.

AspectDetails
CoverageGlobal (Africa, South Asia, Southeast Asia, Latin America, Caribbean)
Buildings1.8+ billion
FormatGeoJSON, CSV
LicenseCC-BY-4.0
Websitesites.research.google/open-buildings

Features:

  • V3 includes confidence scores and plus codes
  • Temporal coverage 2021-2023
  • Optimized for developing regions

Overture Maps Foundation Buildings

Merged and harmonized building data from multiple sources.

AspectDetails
CoverageGlobal
BuildingsGrowing (includes Microsoft, OSM, Esri contributions)
FormatGeoParquet
LicenseCDLA Permissive 2.0
Websiteoverturemaps.org

Features:

  • Building height and floor count attributes
  • Unified schema across sources
  • Cloud-native GeoParquet format
  • Monthly releases

OpenStreetMap Buildings

Community-contributed building footprints.

AspectDetails
CoverageGlobal (varies by region)
Buildings500+ million
FormatOSM PBF, GeoJSON via Overpass
LicenseODbL
Websiteopenstreetmap.org

Access Methods:

Provider Comparison

ProviderCoverageBuildingsFormatHeight DataLicense
Microsoft USUSA129M+GeoJSONNoODbL
Microsoft GlobalGlobal1.2B+GeoJSONNoODbL
Google Open BuildingsGlobal*1.8BGeoJSON/CSVNoCC-BY-4.0
Overture MapsGlobalGrowingGeoParquetYesCDLA
OpenStreetMapGlobal500M+OSM/GeoJSONSometimesODbL

*Google coverage focused on developing regions

Commercial Providers

ProviderDescriptionLinks
RegridBuilding footprints as add-on to parcel dataWebsite
CoreLogic ParcelPointPremium building footprints with attributesWebsite
NearmapAI-derived 3D building modelsWebsite
Pictometry (EagleView)Oblique imagery-derived footprintsWebsite

Access via Cloud Platforms

PlatformAvailable DatasetsLinks
AWS Open DataMicrosoft, OvertureRegistry
Google Earth EngineGoogle Open BuildingsCatalog
Microsoft Planetary ComputerMicrosoft BuildingsWebsite
Source CooperativeOverture, othersWebsite

Usage Examples

R - Download Microsoft Buildings

# download state building footprints
library(sf)
library(httr2)
 
state <- "Colorado"
url <- glue::glue(
 "https://usbuildingdata.blob.core.windows.net/usbuildings-v2/{state}.geojson.zip"
)
 
# download and read
temp <- tempfile(fileext = ".zip")
download.file(url, temp)
buildings <- sf::st_read(unzip(temp))

Python - Query Overture via DuckDB

import duckdb
 
con = duckdb.connect()
con.execute("INSTALL spatial; LOAD spatial;")
 
# query buildings for a bbox
query = """
SELECT *
FROM read_parquet('s3://overturemaps-us-west-2/release/2024-*/theme=buildings/*')
WHERE bbox.xmin > -105.3 AND bbox.xmax < -104.9
  AND bbox.ymin > 39.6 AND bbox.ymax < 39.8
"""
buildings = con.execute(query).fetchdf()

Notes

  • Microsoft provides the best US coverage with consistent quality
  • Overture Maps is emerging as the unified standard with richer attributes
  • Google excels in developing regions with sparse OSM coverage
  • Consider freshness: AI-derived datasets update less frequently than OSM
  • Licensing: Check license requirements for commercial use

Appendix

Created: 2024-12-23 | Modified: 2024-12-23

See Also

(c) No Clocks, LLC | 2024

Graph View

Backlinks