Australia to Vanuatu: A visual exploration of the mapping challenges in Asia...
summary
Mapillary is a street‑level imagery platform that now hosts 401 million images and runs computer vision pipelines that produce semantic segmentation and a 3D point cloud to improve geolocation and detect objects like lanes, traffic signs and trash cans. Governments use it: World Bank teams monitor ~2,000 km of Lao roads, Portland/Hillsboro maps an entire city in days using driver rigs, individuals fill Chiang Mai with POIs, and a camera grant program supplies dashcams and GoPros.
AI-written summary · about
chapters (16)
- 0:00Intro
- 2:22Learning from one another
- 3:01Streetlevel imagery
- 3:28Computer vision
- 4:59History of Open Street Map
- 6:14Pick for Review
- 7:22QGIS Plugin
- 7:50Laos
- 9:23Tagging
- 10:11Chiang Mai
- 11:10Points of Interest
- 11:28Coordinate Capture
- 12:00Portland Oregon
- 12:39Camera Grant Program
- 13:40GoPro
- 13:56Magnetic Mount
related talks
22mIsland Mapping: Measuring Urban Footprints with Mapillary & Open Source Tools
24mRapid GIS data collection using 360 cameras - Case studies in Nauru and Thursday Island
05mLocation Intelligence Down Under...
19mHow we leverage AI to build great maps at massive scale
01h 00mPostGIS in Agribotics
22mState of OpenStreetMap Data in Asia
44mMaps & Data
22mDevelopment and Implementation of Historical and Current Climate Mapping for Vanuatu