Current and Future Trends in Mapping
summary
Lyft handles about 10 million rides/week and covers roughly 95% US population, and mapping work focuses on two paths: an open network for partners plus an internal level 5 self-driving stack. Mapping pipelines use satellite and Street‑level imagery, deep nets that detect street numbers and storefront extents, LiDAR point clouds for 5 cm localization, visual SLAM, heavy human ops (hundreds–thousands of annotators) and crowdsourcing; automation reduces cost but maps require near‑zero error for turn restrictions and address geocoding.
AI-written summary · about
chapters (14)
- 0:00Start
- 1:16About Me
- 2:54Lyft Autonomous Vehicles
- 6:39Maps Data Layer Cake
- 7:44Map Making - Early Days
- 8:24Advent of Commercial Satellite Imagery
- 9:15Tracing over imagery
- 10:20Street Level Imagery
- 12:00Human Operations
- 12:27Automation
- 18:17Ride Sharing
- 20:33Augmented Reality
- 21:38Autonomous Drones (& Flying Cars)
- 21:50Advertising
related talks
17mThe New Meaning of Mapping
16mOrchestrating Ecosystems with Location Intelligence
25mAtlas of Popular Transport: Democratizing mobility data
11mUrban Computing
23mUber and Europe: Partnering to Enable City Transformation I
57mBuilding the best maps in the world: A conversation with leaders from Mapbox and Tripadvisor
16mUsing visualisations for addressing fleet management problems
16m"Putting a New Spin on