Data and description below is from the Scientific Data paper: A rasterized building footprint dataset for the United States authored by Mehdi P. Heris, Nathan Leon Foks, Kenneth J. Bagstad, Austin Troy & Zachary H. Ancona

Abstract

Microsoft released a U.S.-wide vector building dataset in 2018. Although the vector building layers provide relatively accurate geometries, their use in large-extent geospatial analysis comes at a high computational cost. We used High-Performance Computing (HPC) to develop an algorithm that calculates six summary values for each cell in a raster representation of each U.S. state, excluding Alaska and Hawaii: (1) total footprint coverage, (2) number of unique buildings intersecting each cell, (3) number of building centroids falling inside each cell, and area of the (4) average, (5) smallest, and (6) largest area of buildings that intersect each cell. These values are represented as raster layers with 30 m cell size covering the 48 conterminous states. We also identify errors in the original building dataset. We evaluate precision and recall in the data for three large U.S. urban areas. Precision is high and comparable to results reported by Microsoft while recall is high for buildings with footprints larger than 200 m2 but lower for progressively smaller buildings.

Background and Summary

Building footprints are a critical environmental descriptor. Microsoft produced a U.S.-wide vector building dataset in 20181 that was generated from aerial images available to Bing Maps using deep learning methods for object classification2. The main goal of this product has been to increase the coverage of building footprints available for OpenStreetMap. Microsoft identified building footprints in two phases; first, using semantic segmentation to identify building pixels from aerial imagery using Deep Neural Networks and second, converting building pixel blobs into polygons. The final dataset includes 125,192,184 building footprint polygon geometries in GeoJSON vector format, covering all 50 U.S. States, with data for each state distributed separately. These data have 99.3% precision and 93.5% pixel recall accuracy2. Temporal resolution of the data (i.e., years of the aerial imagery used to derive the data) are not provided by Microsoft in the metadata.

Using vector layers for large-extent (i.e., national or state-level) spatial analysis and modelling (e.g., mapping the Wildland-Urban Interface, flood and coastal hazards, or large-extent urban typology modelling) is challenging in practice. Although vector data provide accurate geometries, incorporating them in large-extent spatial analysis comes at a high computational cost. We used High Performance Computing (HPC) to develop an algorithm that calculates six summary statistics (described below) for buildings at 30-m cell size in the 48 conterminous U.S. states, to better support national-scale and multi-state modelling that requires building footprint data. To develop these six derived products from the Microsoft buildings dataset, we created an algorithm that took every single building and built a small meshgrid (a 2D array) for the bounding box of the building and calculated unique values for each cell of the meshgrid. This grid structure is aligned with National Land Cover Database (NLCD) products (projected using Albers Equal Area Conic system), enabling researchers to combine or compare our products with standard national-scale datasets such as land cover, tree canopy cover, and urban imperviousness3.

Locations, shapes, and distribution patterns of structures in urban and rural areas are the subject of many studies. Buildings represent the density of built up areas as an indicator of urban morphology or spatial structures of cities and metropolitan areas4,5. In local studies, the use of vector data types is easier6,7. However, in regional and national studies a raster dataset would be more preferable. For example in measuring the spatial structure of metropolitan areas a rasterized building layer would be more useful than the original vector datasets8.

Our output raster products are: (1) total building footprint coverage per cell (m2 of building footprint per 900 m2 cell); (2) number of buildings that intersect each cell; (3) number of building centroids falling within each cell; (4) area of the largest building intersecting each cell (m2); (5) area of the smallest building intersecting each cell (m2); and (6) average area of all buildings intersecting each cell (m2). The last three area metrics include building area that falls outside the cell but where part of the building intersects the cell (Fig. 1). These values can be used to describe the intensity and typology of the built environment.

Code

Our software is available through U.S. Geological Survey code r...

“Automatically Extracted Buildings” is a raw digital product in vector format created by NRCan. It consists of a single topographical feature class that delineates polygonal building footprints automatically extracted from airborne Lidar data, high-resolution optical imagery or other sources.

This dataset pulls from many different data sources to identify individual building characteristics of all buildings in Boston. It also identifies high-potential retrofit options to reduce carbon emissions in multifamily buildings, using the best available data and assumptions from building experts.

Building characteristics will require on-site verification before an owner can act on them.

Find out more about carbon targets for Boston's existing large buildings.

Download In State Plane Projection Here. The pavement boundaries were traced from aerial photography taken between April 13 and April 26, 2002 and then updated from photography taken between March 15 and April 25, 2018. This dataset should meet National Map Accuracy Standards for a 1:1200 product. Lake County staff reviewed this dataset to ensure completeness and correct classification. In the case of a divided highway, the pavement on each side is captured separately. Island features in cul-de-sacs and in roads are included as a separate polygon.These building outlines were traced from aerial photography taken between April 13 and April 26, 2002 and then updated from successive years of photography. The most recent aerial photography was flown between March 11 and April 12, 2017. This dataset should meet National Map Accuracy Standards for a 1:1200 product. All the enclosed structures in Lake County with an area larger than 100 square feet as of April 2014 should be represented in this coverage. It should also be noted that a single polygon in this dataset could be composed of many structures that share walls or are otherwise touching. For example, a shopping mall may be captured as one polygon. Note that the roof area boundary is often not identical to the building footprint at ground level. Contributors to this dataset include: Municipal GIS Partners, Inc., Village of Gurnee, Village of Vernon Hills.

The Bing Maps team at Microsoft released a U.S.-wide vector building dataset in 2018, which includes over 125 million building footprints for all 50 states in GeoJSON format. This dataset is extracted from aerial images using deep learning object classification methods. Large-extent modelling (e.g., urban morphological analysis or ecosystem assessment models) or accuracy assessment with vector layers is highly challenging in practice. Although vector layers provide accurate geometries, their use in large-extent geospatial analysis comes at a high computational cost. We used High Performance Computing (HPC) to develop an algorithm that calculates six summary values for each cell in a raster representation of each U.S. state: (1) total footprint coverage, (2) number of unique buildings intersecting each cell, (3) number of building centroids falling inside each cell, and area of the (4) average, (5) smallest, and (6) largest area of buildings that intersect each cell. These values are represented as raster layers with 30 m cell size covering the 48 conterminous states, to better support incorporation of building footprint data into large-extent modelling. We also identify errors in the original building dataset where buildings are systematically over- or undercounted, providing further guidance for their use in geospatial analysis. This Project is funded by NASA’s Biological Diversity and Ecological Forcasting program; Award # 80NSSC18k0341

DRAKO specializes in providing Building Footprint Data, offering a unique perspective on spatial analytics and location intelligence. Our data supports businesses in understanding their environments and optimizing their strategies through precise geolocation information.

Building Footprint Data can be used to assess market opportunities, evaluate potential locations, and analyze the competitive landscape. We’re able to provide geographies as addresses, latitude and longitudes; or WKT84 Polygons. Additionally, with our rich dataset, we can provide detailed information about: Business Location, Store Location, and nearby Points of Interest (POI) and Places to ensure you have the insights necessary for informed decision-making. Moreover, we’re also able to reverse geocode data into actionable addresses for you from other geospatial data formats.

Key Features: - Comprehensive mapping of building footprints for accurate spatial analysis Business attributes for each building - Integration with Business Location Data for enhanced market analysis - Access to relevant POIs and Places Data to understand local dynamics - Customizable filters to target specific regions or business types - Able to query by Banner name, ex., Pizza Hut or Walmart - Able to Geofence using foot-print data to create Advertising Audiences using Mobile Location Data

Use Cases: - Site selection and feasibility studies - Market analysis and competitive intelligence - Urban planning and development insights - Real estate assessments and investment strategies - Location-based marketing and outreach

Data Compliance: All of our Building Footprint Data adheres to industry standards for data protection and privacy. We ensure that all data is sourced ethically and responsibly, providing accurate information without compromising user privacy.

Data Quality: DRAKO employs rigorous validation processes to ensure the accuracy and reliability of our Building Footprint Data. Our quality assurance protocols include regular updates and cross-referencing with trusted data sources, ensuring that our information remains current and dependable.

Footprint outlines of buildings in New York City. Please see the following link for additional documentation: https://github.com/CityOfNewYork/nyc-geo-metadata/blob/master/Metadata/Metadata_BuildingFootprints.md For additional resources, please refer to: https://nycmaps-nyc.hub.arcgis.com/search?tags=building&type=feature%2520service%2Cfeature%2520layer

A polygon showing the permanent buildings or structures, including the building footprint. Attributes include the building types, building names, building height etc.

This data shows the digitized building footprints of buildings located within the City of Winchester, Virginia. This data was collected off Eagleview 2017 aerial imagery and was provided to the City after the flight.

Initial Data Capture: Building were originally digitized using ESRI construction tools such as rectangle and polygon. Textron Feature Analyst was then used to digitize buildings using a semi-automated polygon capture tool as well as a fully automated supervised learning method. The method that proved to be most effective was the semi-automated polygon capture tool as the fully automated process produced polygons that required extensive cleanup. This tool increased the speed and accuracy of digitizing by 40%.Purpose of Data Created: To supplement our GIS viewers with a searchable feature class of structures within Ventura County that can aid in analysis for multiple agencies and the public at large.Types of Data Used: Aerial Imagery (Pictometry 2015, 9inch ortho/oblique, Pictometry 2018, 6inch ortho/oblique) Simi Valley Lidar Data (Q2 Harris Corp Lidar) Coverage of Data:Buildings have been collected from the aerial imageries extent. The 2015 imagery coverage the south county from the north in Ojai to the south in thousand oaks, to the east in Simi Valley, and to the West in the county line with Santa Barbara. Lockwood Valley was also captured in the 2015 imagery. To collect buildings for the wilderness areas we needed to use the imagery from 2007 when we last flew aerial imagery for the entire county. 2018 Imagery was used to capture buildings that were built after 2015.Schema: Fields: APN, Image Date, Image Source, Building Type, Building Description, Address, City, Zip, Data Source, Parcel Data (Year Built, Basement yes/no, Number of Floors) Zoning Data (Main Building, Out Building, Garage), First Floor Elevation, Rough Building Height, X/Y Coordinates, Dimensions. Confidence Levels/Methods:Address data: 90% All Buildings should have an address if they appear to be a building that would normally need an address (Main Residence). To create an address, we do a spatial join on the parcels from the centroid of a building polygon and extract the address data and APN. To collect the missing addresses, we can do a spatial join between the master address and the parcels and then the parcels back to the building polygons. Using a summarize to the APN field we will be able to identify the parcels that have multiple buildings and delete the address information for the buildings that are not a main residence.Building Type Data: 99% All buildings should have a building type according to the site use category code provided from the parcel table information. To further classify multiple buildings on parcels in residential areas, the shape area field was used to identify building polygons greater than 600 square feet as an occupied residence and all other buildings less than that size as outbuildings. All parcels, inparticular parcels with multiple buildings, are subject to classification error. Further defining could be possible with extensive quality control APN Data: 98% All buildings have received APN data from their associated parcel after a spatial join was performed. Building overlapping parcel lines had their centroid derived which allowed for an accurate spatial join.Troubleshooting Required: Buildings would sometimes overlap parcel lines making spatial joining inaccurate. To fix this you create a point from the centroid of the building polygon, join the parcel information to the point, then join the point with the parcel information back to the building polygon.

The Open Database of Buildings (ODB) is a collection of open data on buildings made available under the Open Government License - Canada. The ODB brings together 530 datasets originating from 107 government sources of open data. The database aims to enhance access to a harmonized collection of building features across Canada.

Microsoft Buildings Footprints with Heights from service: https://services.arcgis.com/P3ePLMYs2RVChkJx/arcgis/rest/services/MS_Buildings_Training_Data_with_Heights/FeatureServer (restrictions, do not use)Source: Approx. 9.8 million building footprints for portions of metro areas in 44 US States in Shapefile format.Microsoft recently released a free set of deep learning generated building footprints covering the United States of America. As part of that project Microsoft shared 8 million digitized building footprints with height information used for training the Deep Learning Algorithm. This map layer includes all buildings with height information for the original training set that can be used in scene viewer and ArcGIS pro to create simple 3D representations of buildings. Learn more about the Microsoft Project at the Announcement Blog or the raw data is available at Github.Click see Microsoft Building Layers in ArcGIS Online.Digitized building footprint by State and CityAlabamaGreater Phoenix City, Mobile, and MontgomeryArizonaTucsonArkansasLittle Rock with 5 buildings just across the river from MemphisCaliforniaBakersfield, Fresno, Modesto, Santa Barbara, Sacramento, Stockton, Calaveras County, San Fran & bay area south to San Jose and north to CloverdaleColoradoInterior of DenverConnecticutEnfield and Windsor LocksDelawareDoverFloridaTampa, Clearwater, St. Petersburg, Orlando, Daytona Beach, Jacksonville and GainesvilleGeorgiaColumbus, Atlanta, and AugustaIllinoisEast St. Louis, downtown area, Springfield, Champaign and UrbanaIndianaIndianapolis downtown and Jeffersonville downtownIowaDes MoinesKansasTopekaKentuckyLouisville downtown, Covington and NewportLouisianaShreveport, Baton Rouge and center of New OrleansMaineAugusta and PortlandMarylandBaltimoreMassachusettsBoston, South Attleboro, commercial area in Seekonk, and SpringfieldMichiganDowntown DetroitMinnesotaDowntown MinneapolisMississippiBiloxi and GulfportMissouriDowntown St. Louis, Jefferson City and SpringfieldNebraskaLincolnNevadaCarson City, Reno and Los VegasNew HampshireConcordNew JerseyCamden and downtown Jersey CityNew MexicoAlbuquerque and Santa FeNew YorkSyracuse and ManhattanNorth CarolinaGreensboro, Durham, and RaleighNorth DakotaBismarckOhioDowntown Cleveland, downtown Cincinnati, and downtown ColumbusOklahomaDowntown Tulsa and downtown Oklahoma CityOregonPortlandPennsylvaniaDowntown Pittsburgh, Harrisburg, and PhiladelphiaRhode IslandThe greater Providence areaSouth CarolinaGreensville, downtown Augsta, greater Columbia area and greater Charleston areaSouth Dakotagreater Pierre areaTennesseeMemphis and NashvilleTexasLubbock, Longview, part of Fort Worth, Austin, downtown Houston, and Corpus ChristiUtahSalt Lake City downtownVirginiaRichmondWashingtonGreater Seattle area to Tacoma to the south and Marysville to the northWisconsinGreen Bay, downtown Milwaukee and MadisonWyomingCheyenne

This vector polygon dataset represents the building features in Yosemite National Park. This dataset utilizes the updated NPS Building Spatial Data Standard dated 12/15/20217. There are ongoing efforts to improve the spatial and attribute information of the buildings.Initial polygons were digitized from various sources with unknown provenance, likely satellite imagery or CAD files. Many existing building polygons were added or updated from 3D building footprints covering the 2019 Yosemite National Park 3DEP Lidar project area. New buildings are COGOed where possible, otherwise digitized from satellite imagery or extracted from DWG files on an as-needed basis. In 2025 a volunteer georeferenced old maps of the park and digitized some buildings that have been removed since the time of the maps' making. Information about how each building polygon was created is in the Map Method, Map Source, and Source Date fields.Polygons are meant to represent the building footprint, though there are still buildings represented with roof outlines, particularly private residences and others digitized from satellite imagery. Buildings with more than one FMSS Locations are split to delineate the multiple assets, even though the footprint is connected. When two or more footprints share a roof they are represented with multi-part polygons that represent the foundations of the buildings.Attributes in this dataset include identifier fields (building name and label fields, FMSS Location ID, and various other ID fields), the current state of the structure (Status and Is Extant fields), classification (Functional and Facility Use fields as well as Seasonality, Building Code, and Building Type), as well as record level metadata fields. Efforts by various staff members over the years have standardized and corrected many of these fields for most of the buildings, but inaccuracies remain.This dataset is meant for both public and internal use, with sharing status described in the Public Map Display and Data Access fields. Non-extant buildings are marked as No Public Map Display but remain a part of the dataset to provide insight into what the park used to look like.IRMA Data Store Reference

Buildings

## Overview

Buildings is a dataset for instance segmentation tasks - it contains Buildings annotations for 598 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

NYS Building Footprints - metadata info:The New York State building footprints service contains building footprints with address information. The footprints have address point information folded in from the Streets and Address Matching (SAM - https://gis.ny.gov/streets/) address point file. The building footprints have a field called “Address Range”, this field shows (where available) either a single address or an address range, depending on the address points that fall within the footprint. Ex: 3860 Atlantic Avenue or Ex: 32 - 34 Wheatfield Circle Building footprints in New York State are from four different sources: Microsoft, Open Data, New York State Energy Research and Development Authority (NYSERDA), and Geospatial Services. The majority of the footprints are from NYSERDA, except in NYC where the primary source was Open Data. Microsoft footprints were added where the other 2 sources were missing polygons. Field Descriptions: NYSGeo Source : tells the end user if the source is NYSERDA, Microsoft, NYC Open Data, and could expand from here in the futureAddress Point Count: the number of address points that fall within that building footprintAddress Range : If an address point falls within a footprint it lists the range of those address points. Ex: if a building is on a corner of South Pearl and Beaver Street, 40 points fall on the building, and 35 are South Pearl Street it would give the range of addresses for South Pearl. We also removed sub addresses from this range, primarily apartment related. For example, in above example, it would not list 30 South Pearl, Apartment 5A, it would list 30 South Pearl.Most Common Street : the street name of the largest number of address points. In the above example, it would list “South Pearl” as the most common street since the majority of address points list it as the street. Other Streets: the list of other streets that fall within the building footprint, if any. In the above example, “Beaver Street” would be listed since address points for Beaver Street fall on the footprint but are not in the majority.County Name : County name populated from CIESINs. If not populated from CIESINs, identified by the GSMunicipality Name : Municipality name populated from CIESINs. If not populated from CIESINs, identified by the GSSource: Source where the data came from. If NYSGeo Source = NYSERDA, the data would typically list orthoimagery, LIDAR, county data, etc.Source ID: if NYSGeo Source = NYSERDA, Source ID would typically list an orthoimage or LIDAR tileSource Date: Date the footprint was created. If the source image was from 2016 orthoimagery, 2016 would be the Source Date. Description of each footprint source:NYSERDA Building footprints that were created as part of the New York State Flood Impact Decision Support Systems https://fidss.ciesin.columbia.edu/home Footprints vary in age from county to county.Microsoft Building Footprints released 6/28/2018 - vintage unknown/varies. More info on this dataset can be found at https://blogs.bing.com/maps/2018-06/microsoft-releases-125-million-building-footprints-in-the-us-as-open-data.NYC Open Data - Building Footprints of New York City as a polygon feature class. Last updated 7/30/2018, downloaded on 8/6/2018. Feature Class of footprint outlines of buildings in New York City. Please see the following link for additional documentation- https://github.com/CityOfNewYork/nyc-geo-metadata/blob/master/Metadata/Metadata_BuildingFootprints.mdSpatial Reference of Source Data: UTM Zone 18, meters, NAD 83. Spatial Reference of Web Service: Spatial Reference of Web Service: WGS 1984 Web Mercator Auxiliary Sphere.

Dataset

This dataset was created by jamesnb

Contents

Mipt Buildings

## Overview

Mipt Buildings is a dataset for object detection tasks - it contains Buildings annotations for 620 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

zhangtao-whu/buildings dataset hosted on Hugging Face and contributed by the HF Datasets community