Urban Artificial Intelligence in Amsterdam
summary
City projects use concrete ML pipelines: a text+photo classifier routes ~300,000 annual reports into 130 service categories and raises customer satisfaction +40%; cheap €200 phones on ~1,200 vehicles collect millions of street images, with ~25,000 labeled images used to train the open-source object detection kit (odk.ai) to spot mattresses, trash types and furniture for targeted pickup. Systems remove faces/license plates, run crowd counting for park load balancing, and process multi-terabyte LiDAR point clouds to map 125k streetlights and 250k trees.
AI-written summary · about
chapters (8)
- 0:00Start
- 5:19Image Classification
- 6:11Examples of Ai Machine Learning Projects
- 6:513d Point Cloud
- 13:50Privacy
- 16:13Crowd Counting
- 17:15Point Clouds
- 20:23Object Detection
related talks
10mMapping sidewalks and street furniture with AI and computer vision
19mHow we leverage AI to build great maps at massive scale
32mIsolated Artificial Intelligence: Now or Tomorrow?
46mML Use Cases at Think Big Analytics
46mData Discussion - AI for Good
28mGeoAI: Evolution of AI-Based Tree/Vegetation Detection Models Using Imagery
35mDriving Societal Change Through AI Innovation
06mAssisted Road Surface Tagging using Machine Learning