Mapillary has released a 3D visualization tool online. Behind-the-scenes, Mapillary, a community-based mapping photo app, has been using machine vision to stitch together its users' photos and 3D construct every location their community photographs. The goal is to improve positioning of each photo beyond its GPS coordinates, but 3D navigation also provides an ethereal viewing experience.

Now Mapillary is giving users the power to virtually explore their own photos like never before. Underlying 3D data can now be viewed by clicking the “show point clouds” option in the sidebar of a web browser. Full support for panoramas has also been added, enabling users to move seamlessly between regular photos and panoramas in 3D, giving them a smooth and other-worldly viewing experience.

Mapillary consumers have taken more than 40 million photos to date. They are stitched together with computer vision, a type of artificial intelligence that extracts information from images. For every single photo uploaded, Mapillary can automatically match features to the ones in photos from the same geographic vicinity so that computing how the images relate to each other and how to navigate from one to the other can be computed. This is how users navigate between photos in the Mapillary app and web browser. Taking it one step further, Mapillary can also recover a 3D model of every area photographed.

By having behind-the-scenes access to the 3D navigation, members can now explore the underlying 3D data in the web view, which addresses common questions about new transitions on the photo navigation and accurately positioning randomly connected photos.

Examples include:

  • The 3D data from one person walking across the courtyard at Alcazar, Sevilla. Camera positions and white rectangle frames can be seen.
  • The underlying 3D model of a cemetery building from one person walking back and forth across the open space in front.

About Mapillary

Mapillary is a service for crowdsourcing map photos. Using simple tools like smartphones or action cameras anyone can collect photos that are combined into a collective street level photo view. The service provides apps for iOS, Android and Windows Phone, together with tools for using action cameras like GoPro and Garmin Virb. Mapillary also provides a set of APIs so that others can build solutions using the photos and the data extracted. There is also a viewer and widget that anyone can embed.