Head Image Caption: Street level view of 3D-reconstructed Chelsea, Manhattan
Historians and nostalgic residents alike take an interest in how cities were constructed and how they developed — and now there’s a tool for that. Google AI recently launched the open-source browser-based toolset “rǝ,” which was created to enable the exploration of city transitions from 1800 to 2000 virtually in a three-dimensional view.
Google AI says the name rǝ is pronounced as “re-turn” and derives its meaning from “reconstruction, research, recreation and remembering.” This scalable system runs on Google Cloud and Kubernetes and reconstructs cities from historical maps and photos.
There are three main components to the toolset.
Warper is a crowdsourcing platform,where users can upload photos of historical print maps and georectify them to match real world coordinates. These can then be converted to an OpenStreetMap (OSM) vector format with an Editor app.
A time-dimension slider on a temporal map server allows users to move between time periods to see how city maps change.
The team also built a 3D experience platform with the support from state-of-art deep learning models such as region-based convolutional neural networks (RCNN), semantic segmentation model DeepLab, and a specifically designed neural network. The 3D reconstruction pipeline recognizes and reconstructs all individual constituent components separately based on their categories. The detailed 3D structures are then merged with the coarse one for the final 3D mesh. This 3D platform is still under construction, and the project is welcoming historical map contributions.
Google AI says it hopes rǝ will provide “a rewarding ‘time travel’ experience for both research and entertainment purposes,” and so far the open-source tool suite is being positively received by the research community and others on Twitter.
The associated paper Kartta Labs: Collaborative Time Travel is available on arXiv.
Reporter: Reina Qi Wan | Editor: Michael Sarazen; Yuan Yuan