Mapillary continues to be an important tool for editing OpenStreetMap. Over 500,000 changesets using Mapillary have been added in 2020. These edits have come from community members, corporate editors, and other organized groups like our friends at YouthMappers and the Humanitarian OpenStreetMap Team. Finding creative uses for street-level imagery is a defining strength of Mapillary users everywhere, from mapping bicycle infrastructure to utility poles to speed limits.

Our team has been focused on improving Mapillary functionality within OSM’s iD editor. This will make it easier than ever to reference street-level imagery and computer vision extracted data when improving the map. The latest release of OSM iD includes many new Mapillary features, and today we’ll give a brief overview of what these are and how you can use them.

1. Linking to Mapillary images in the URL

In previous releases of OSM iD, a new feature allowed the photo overlay layer to be activated from the URL, with a link like this that would take the user to Hawaii:

"https://www.openstreetmap.org/edit#map=18/21.4576/-157.8355&photo_overlay=mapillary"

Now, it is possible to give a link to the iD Editor with even more detail, including the image key you want to reference. This makes it possible to view an image in another app, then quickly open the iD Editor using that same image without having to search for it again. In the same location as before, we can now use this URL:

"https://www.openstreetmap.org/edit#map=18/21.4576/-157.8355&photo_overlay=mapillary&photo=mapillary/vYZR9VxrB0OrXD5RmAPcOA"

When the map loads, the Mapillary layer is on, and the photo also loads. You’re ready to start editing at this specific location!

2. See image of map features without image layer

The Mapillary image layer can obstruct the view, with the green dots and lines covering up roads and points underneath. When you’re just adding data from the Mapillary map features layer, you can keep the images layer off, and now just click on an object to load and see the location of an image that contains it. Notice below how after clicking a utility pole, you can see the gold location indicator for an image, as well as boxes around the other map features visible in the image.

3. Filter images by date and username

Mapillary’s website has had filters for a while, but this functionality has now come to iD Editor. Did you capture imagery last week in your local national park? Are you looking for images captured by your fellow community members? You can now set a filter to focus only on the images you want, making it easier to reference these images for map editing.

To activate the filters, you can simply scroll down in the Map Data menu (F hotkey), and below the checkboxes which already allow filtering out flat or panoramic 360 images, you can enter the date or username (or multiple usernames, comma separated), and the filter will automatically be applied.

These filters can also be included in the URL, along with the image key. For the same filter as used above, try this in the URL:

"https://www.openstreetmap.org/edit#map=20/47.52405/18.99357&photo_overlay=mapillary&photo=mapillary/Jybl_gbNh0tPO1KXV2SYdQ&photo_dates=2020-03-01_2020-06-10&photo_username=chrisbeddow"

4. Bug Fixes and In Case You Missed It

Beyond these new features, a few improvements have been made under the hood. These include fixing disappearing Mapillary viewer controls and improving performance of the map feature tiles.

In case you didn’t notice some previous new features, a few to keep in mind include: - Click and drag the upper right corner of the Mapillary viewer to resize - Map features layer is available globally - Click a map feature or traffic sign icon to load an image viewing it

A special thanks goes out to our talented front-end engineer, Nikola Pleša, who made these latest features a reality, and to Quincy Morgan who has been dedicated to keeping the iD Editor running smoothly.

Going forward, we plan to continue improving the Mapillary tools in both iD and JOSM, especially now that our new home is with Facebook’s Spatial Computing team where a focus on OpenStreetMap is key. If you have bugs, please report them in the OSM iD Github repository. We also welcome feature requests, use cases, or any desire to chat and show us how you’re using Mapillary, so email us at osm@mapillary.com. We hope these new features will expand the possibilities of using Mapillary with OpenStreetMap, and look forward to checking in with the community on future exciting releases.

/Chris