$ show microsoft/GlobalMLBuildingFootprints
This repository provides access to Microsoft's global building footprint dataset (1.4B buildings) derived from Bing Maps satellite imagery. The core deliverable is a massive, publicly hosted dataset of building polygons in GeoJSONL format, with periodic updates. The repository itself contains only documentation, a Jupyter notebook demonstrating data download and spatial filtering, and a few utility scripts for converting compressed GeoJSONL files to GIS-friendly formats. There is no transformation pipeline or model training code present; the actual building extraction is done externally. The code is simple, well-documented, and parameterized for reuse, but lacks dependency management and environment definitions. The primary value is the dataset itself, with the scripts serving as examples for accessing and processing the data.
The LLM identified 3 files worth highlighting for reuse:
A clean, parameterized Python script that downloads building footprints for a specified country from the dataset-links.csv, reads line-delimited GeoJSON, converts geometry objects, and saves as a GeoJSON file. Demonstrates a reusable pattern for accessing and transforming geospatial data from a remote index, easily adaptable to other tiled datasets.
A robust script for splitting a large line-delimited GeoJSONL file into smaller chunks, with configurable feature count per file. Handles edge cases (assertions for file existence, no overwrites) and is memory-efficient. This pattern is reusable for any large line-oriented dataset where chunking is needed.
A well-structured Jupyter notebook that demonstrates the full workflow: define an area of interest, determine intersecting quadkeys using mercantile, download building footprints from the dataset-links index, filter geometries within the AOI, and save as GeoJSON. Includes clear cell organization and uses tempfile for intermediate files. This pattern is reusable for any tiled geospatial dataset with a similar index.