Skip to main content
Uber logo

Schedule rides in advance

Reserve a rideReserve a ride

Schedule rides in advance

Reserve a rideReserve a ride
Data / ML, Engineering

Taking City Visualization into the Third Dimension with Point Clouds, 3D Tiles, and deck.gl

October 29, 2019 / Global
Featured image for Taking City Visualization into the Third Dimension with Point Clouds, 3D Tiles, and deck.gl
Figure 1. Using Tile3DLayer and deck.gl, we can render Melbourne, Australia’s 355 million data points and achieve an unprecedented level of geospatial detail.
Figure 3. Using 3D Tiles, big geospatial data sets are divided into a spatial hierarchy. Child tiles cover smaller volumes and typically do not need need to be loaded when visualizing at the full data set, but provide additional detail when zooming in. Figure used with permission from Cesium.
Figure 4. 3D Using loaders.gl, we can incrementally load 3D tiles during zooming to keep visualization quality high even while rendering monumentally large data sets.
Figure 5. The loaders.gl architecture centers around the CesiumJS-derived spatial hierarchy traversal algorithm, which is applicable across tile formats, as well as parsers for the individual tiles in each format. New renderer integrations can also be added independently.
Figure 6. We can use loaders.gl to incrementally load 3D tiled point clouds in deck.gl.
Ib Green

Ib Green

Ib Green is a staff software engineer on Uber’s Visualization team.

Chris Gervang

Chris Gervang

Chris is a senior software engineer on Uber Elevate’s System Simulation platform team. He develops Flux, Elevate’s operation and network design framework, and Hubble.gl, an animation library for 3D visualization.

Ian Villa

Ian Villa

Ian is the System Simulation Lead at Uber Elevate. He works with the team to develop integrated market simulations of networks, skyports, and aircraft across Uber’s aviation products. Previously, Ian was the technical lead for aircraft partnerships at Uber Elevate.

Posted by Ib Green, Chris Gervang, Ian Villa