We already have beautiful 3D maps, but the public lacks useful 3D maps. The photorealism of Google Earth and Apple Maps doesn't help us navigate indoors, explore building schematics for repairs and studies, etc.
The idea is to apply border detection and line drawing algorithms to existing photorealistic imagery (on the left of the Fig. 1.) so we can create the exoskeletons of buildings (middle of Fig. 1.) which can then be filled in collaboratively using floor plans and schematics from various sources. (public records would be particularly useful)
Floorplans will prove to be highly useful in helping people navigate the complex corridors of skyscrapers and other dense urban areas.
X-Ray Vision with Augmented Reality
It'd be rad to use these highly detailed 3D schematics to give people x-ray vision through buildings with handheld devices or headsets.
I made a mockup of what this could look like while I was at a coffee shop in Deep Ellum, Dallas. Just illustrates how dope it'd be.
Reduced Data Transmission
An additional benefit to the wireframe 3D imagery, is that if it only draws the borders of the buildings, it'll require significantly less data than the current photorealistic buildings.
However, complex internal 3D wireframes might bring resource usage back to parity. so whatever, I guess.
Google, please build this. Beyond begging the tech giants, I'm not sure how the FOSS community could go about building this, since we don't have access to the 3D imagery through APIs. So tweet article this at your friends who work at Google 😉
Fig. 1. - Note Feb 4, 2018
Fig. 2. - Note Jan 8, 2018
Fig. 3. - Note Feb 4, 2018
Fig. 4. - Note Feb 8, 2018 pg 5/10