Exciting Progress in Kiama, NSW! Sharing another milestone in our Kiama project, made possible through collaboration with Kiama Municipal Council, UNSW, and Mapillary as part of the NSW SmartCity Innovation Challenge!
In just a few days, we successfully mapped the entire footpath network in Kiama, utilizing over 2,000 footpath view images. Prior to this, the available footpath network and accessibility data on OpenStreetMap for Kiama was very limited - almost virtually non-existent we can say. Our AI-detected pedestrian infrastructure map data is then overlaid onto OpenStreetMap through Meta's Rapid Editor. A team of humans then validates, improves the quality of the data where needed, and finally publishes the detailed map data on OSM for a broader impact and use by the community!
How does our technology work?
Our DeepWalk™ AI stack automatically identifies and detects objects and features from 360° footpath view (walking person's view) imagery and projects them onto a map with georeferenced points, lines, and polygons. The output includes detailed data on the pedestrian infrastructure (e.g. presence of the footpath, estimated width of the footpath, surface material, presence of weather protection from tree canopies and/or awnings, different assets like benches, bins, and poles as well as accessibility features like lowered curb cuts and marked crosswalks).
Collaboration with Mapillary: Contributing back to OpenStreetMap
You can view all our 360° sidewalk view imagery from Kiama on Mapillary in addition to our own map browser. A huge shoutout to the amazing Mapillary team for their unwavering collaborative spirit. Though we missed being at State of the Map Europe in Antwerp this year to share our work with the OpenStreetMap community, we'll present some of our work at the State of the Map Asia in Bangkok later this week (16-18 November). Collaborative mapping is teamwork in action, navigating a range of geospatial data challenges. AI-assisted mapping acts as a turbo boost, ramping up work efficiency, especially for historically under-represented pedestrian infrastructure on digital maps.
Need your city footpath network mapped?
If you'd like the pedestrian network and assets in your city mapped next, drop us a line to chat.