This talk will cover some of the following areas:
1. An overview of Mapillary.
2. Who is contributing and some interesting case studies.
3. Some of the recent updates including 2.0 mobile apps, NeRFs, and improved upload.
4. How to contribute including cameras, upload tools, and best practices.
5. How to download data using Mapillary's web interface and API.

Since Mapillary launched in 2013, over 2 billion images have been contributed from places as far afield as Antarctica and Zimbabwe. Images can be uploaded from any device that creates geotagged images, from affordable smartphones to commercial grade 360° cameras.

Every image is processed with computer vision to recreate the world in 3D and extract features that are useful for map making. These capabilities have attracted all sorts of map builders including advocates for pedestrian safety, humanitarian agencies, state and local transportation departments, OpenStreetMap contributors, ridesharing companies and more.

In this talk we’ll recap Mapillary for those that are less familiar, sharing some more recent case studies to help crystalize the utility of street-level imagery. We’ll then cover some of the platform changes of 2024, including the launch of the revised mobile apps (2.0). This leads into our latest recommendations for how to capture and upload street-level imagery effectively. We’ll conclude with a look at how you can download map features using the web interface and Python tools.