November 2023

Conference Paper

Building-Level Comparison of Microsoft and Google Open Building Footprints Datasets

By:
Gonzales, Jack J
Page Number:
35:1-35:6
Issue Number:
1
Book Title:
Proceedings of the 12th International Conference on Geographic Information Science (GIScience 2023)
Publication Date:
November 2023
Publisher Location:
Dagstuhl Publishing, Germany
Conference Name:
Twelfth International Conference on Geographic Information Science (GIScience 2023)
Conference Location:
Leeds, United Kingdom
Conference Sponsor:
GIScience
View DOI Listing:
https://doi.org/10.4230/LIPIcs.GIScience.2023.35

Abstract

Large-scale datasets of building footprints are a crucial source of information for a variety of efforts. In 2023, the general public benefits from open access to multiple sources of building footprints at the country scale or larger, such as those produced by Microsoft and Google. However, none of the available datasets have attained complete global coverage, and researchers and analysts may need to combine multiple sources to assemble a complete set of building footprints for their area of interest or choose between overlapping sources, requiring an understanding of the differences between different building sources. This paper presents a method to closely examine the quality of different building footprint sources by matching corresponding buildings across datasets, using building footprints in Ethiopia published by Microsoft and Google as an example set.


Related Researchers