Building Heights: From Open USGS lidar to open Overture maps
summary
Uses USGS 3DEP QL1/QL2 lidar and conflated footprints from OpenStreetMap, Esri and Microsoft to estimate building heights at scale, releasing results via Overture. Pipeline cookies out points inside footprints, estimates roof and terrain elevations, and tags OSM IDs; April release covers 34,000 km², 6 million buildings, cumulative 87,000 km², 16 million buildings. QA uses stratified Zoom‑15 sampling, control‑point annotation from oblique imagery and validation showing majority <1 m error. Known issues: missing classifications, ground bias, and complex building parts; codebase planned to be opened.
AI-written summary · about
chapters (8)
- 0:00Introduction and goals
- 1:34Project scope and overview
- 2:27Use cases and data sources
- 5:12Data coverage and releases
- 7:08Technical methodology
- 9:18Quality assurance strategies
- 11:39Challenges and future work
- 15:26Closing and Q&A
related talks
48mEmpowering AI Innovation Through Open Geospatial Ecosystems
01h 41mOpenStreetWhat? Mapping The World With Open Data
13mUsing open source mapping tools to understand indoor navigation routes...
22mIsland Mapping: Measuring Urban Footprints with Mapillary & Open Source Tools
22mUnderstanding our cities with FOSS4G
17mLaying the Foundations for a National Building Outlines Dataset
23mDigital Earth Models
25mWhat's new in Mapbox Search: BUILD 2025 highlights