Mapillary Blog
blog.mapillary.com · Company blog · on HN since 2014
The Mapillary blog covers crowdsourced street-level imagery, 3D point clouds, computer vision and the company's 2020 acquisition by Facebook.
What HN saysHN links it for the Facebook acquisition news and Mapillary's role in OpenStreetMap; commenters value its image contributions but dislike the loss of contributor attribution.
34HN people shared it
17linked it in comments
41stories submitted
1.2Kstory points
1 year agolast shared
2014people per year2026
Most-shared posts
- 1Facebook acquires Mapillary5 people · 237 points · first shared 2020 · 63 comments on HN
- 2Pointclouds3 people · first shared 2015
- 3The World’s Largest Street-Level Imagery Dataset for Teaching Machines to See2 people · 140 points · first shared 2017 · 7 comments on HN
- 4Denser 3D Point Clouds in OpenSfM2 people · 117 points · first shared 2016 · 36 comments on HN
- 5Turn a Tesla into a mapping vehicle with Mapillary2 people · 70 points · first shared 2020 · 28 comments on HN
- 6Mapillary transforms crowd sourced streetview into interactive 3D2 people · 22 points · first shared 2014 · 4 comments on HN
- 7Traffic Sign Recognition2 people · 10 points · first shared 2015 · discussion on HN
- 8Mapillary time travel anywhere where there is overlap2 people · 3 points · first shared 2016 · discussion on HN
- 9The street view layer of openstreetmap2 people · 2 points · first shared 2016 · 1 comments on HN
- 10Compare places over time (using SfM)2 people · 1 points · first shared 2016 · discussion on HN
- 11100M Photos – Geotagged, Connected, and Available for All1 people · 228 points · first shared 2016 · 31 comments on HN
- 12Mapillary in the OpenStreetMap iD Editor1 people · 142 points · first shared 2014 · 24 comments on HN
- 13Mapillary Now Able to Recognize and Label Objects in the Wild1 people · 39 points · first shared 2016 · 5 comments on HN
- 14Mapillary iD Editor for OpenStreetMap1 people · 34 points · first shared 2014 · 3 comments on HN
- 15Capturing Antarctica1 people · 25 points · first shared 2015 · discussion on HN
What HN people wrote when they linked it
mtmail · 2023-09-19
I cycled every road, every path in a 5km radius. OpenStreetMap back then was empty and with a GPS device recording I filled the map. At least in Europe the OpenStreetMap coverage is so high you won't be able find a blank map area (canvas?) any more. It was fun, the last 20% felt like work.
Other people have done it and on larger scale. This guy for example managed 4500km capturing 600.000 images in Brussels https://blog.mapillary.com/update/2023/07/28/Mapping-Brussels-By-Bike.…
toomuchtodo · 2022-12-20
Mapillary was acquired by Facebook/Meta FYI. Might not matter, surfacing in case it does.
https://blog.mapillary.com/news/2020/06/18/Mapillary-joins-Facebook.ht…
jakecopp · 2022-12-12
Yep I've contributed a fair few Mapillary images! I like that OpenStreetMap integrates with it well; I don't like that it no longer shows _who_ contributed an image (whether in the OSM viewer or the Mapillary viewer).
I believe OpenDroneMap uses OpenSfM which Mapillary open sourced:
https://blog.mapillary.com/update/2016/10/31/denser-3d-point-clouds-in…
> OpenSfM is also used internally by the OpenDroneMap project. OpenDroneMap creates 3D models and orthophotos from drone imagery. It includes all the steps of the pipeline so one can go from images to 3D models in one command. It uses OpenSfM to get the camera positions, and can now use it to compute the dense point clouds as well.
jmkb · 2020-06-21
Admins, consider changing the link to https://blog.mapillary.com/news/2020/06/18/Mapillary-joins-Facebook.ht… -- the current link implies that this is directly related to OpenStreetMap, when in truth it's just a blog ("diary") of a Mapillary employee hosted under his OSM user account.
Kiro · 2020-06-21
This is not the announcement post but in the very first sentence it says "Today we announced that Mapillary has joined Facebook", linking to https://blog.mapillary.com/news/2020/06/18/Mapillary-joins-Facebook.ht…
Seems pretty clear to me.
Piskvorrr · 2019-10-04
This, on a software discussion board. So many horrible errors in such a small package! I sort of have ~3 different map+nav systems, and they won't even agree with each other what the situation is, never mind the Real World (tm). Speed limits specifically: they are in agreement about 80% of the time.
Machine vision, triple meh. Even with strictly standardized set of easily recognizable shapes, this is useful for post facto mapping ( https://blog.mapillary.com/tutorials/2017/08/21/mapping-traffic-signs-… ), but nowhere near realtime without false negatives.
akavel · 2018-07-27
As to humanitarian use for OSM and how anyone can contribute by small tasks (I've tried it myself, was surprisingly exciting), enabling faster and more accurate response and travel to people in need, see:
https://wiki.openstreetmap.org/wiki/Humanitarian_OSM_Team
https://wiki.openstreetmap.org/wiki/OSM_Tasking_Manager
As to work in progress on detection of features and automatic conversion to maps, I've recently found the below website:
https://blog.mapillary.com/update/2016/07/15/mapillary-osm.html
maxerickson · 2018-02-13
There's multiple efforts along those lines.
Mapillary provides sign detections to OSM:
https://blog.mapillary.com/product/2017/02/06/towards-global-traffic-s…
Telenav is doing similar through their OpenStreetCam project (but no good link). I'm not entirely sure, but I think they have the goal of running the detection locally.
Mapbox recently used Bing Streetside imagery to detect turn restrictions (which are important for high quality routing):
https://www.openstreetmap.org/user/daniel-j-h/diary/43242
Mapbox and Telenav also compare their incoming position data with OSM and flag discrepancies (missing roads, turn restrictions again, stuff like that).
mcbits · 2016-11-15
It looks like an attempt to warp between the images, but it's not matching up the features correctly. It works a little better in the "around the world" examples. Some parts just fade or have the wonkywarp effect, but it produces a smooth transition in other parts. http://blog.mapillary.com/update/2016/11/15/around-the-world-in-100-mi…
ungzd · 2016-10-18
Takeout is available (https://takeout.google.com/settings/takeout). There's similar modern service for geotagged photos called Mapillary where you can migrate photos from Panoramio http://blog.mapillary.com/update/2016/10/10/migrating-from-panoramio-a…
petern · 2016-08-10
I can be, since the time perspective is maintained and can be filtered out, see http://blog.mapillary.com/update/2016/03/21/mapillary-js-slider.html
eliaspro · 2015-08-16
You might be interested in OpenDroneMap [1] and Mapillary's OpenSfM [2],[3]
[1] http://opendronemap.github.io/odm/pages/about.html
[2] https://github.com/mapillary/OpenSfM
[3] http://blog.mapillary.com/update/2014/11/14/opensfm.html
a3_nm · 2014-10-23
Maybe the relevant section is left over from the previous license, which was CC-BY-NC (http://blog.mapillary.com/update/2014/03/17/mapillary-goes-creative-co… Alternatively, IANAL but I imagine that the individual photos could still be CC-BY-SA (if, e.g., someone else redistributes them) but that the service itself cannot be used for commercial purposes. As bulk exports are prohibited (2.1) and dumps are not provided, however, this would mean that the data is CC-BY-SA in theory but hard to reuse commercially except for individual pictures.
yellowbkpk · 2014-08-17
They have Android and iOS apps. They also have an API [0] and several blog posts about how to use it to add photos from other types of devices [1].
[0] http://www.mapillary.com/developer.html
[1] http://blog.mapillary.com/technology/2014/07/21/upload-scripts.html
pella · 2016-03-17
The project ( Dar Es Salaam mapping ) is using OpenStreetMap technology and Data. And related to the Humanitarian OpenStreetMap Team ( https://hotosm.org/ )
- http://ramanihuria.org/resources/
- http://ramanihuria.org/about/about-openstreetmap-osm/
- http://ramanihuria.org/news/
"Dar Ramani Huria is a community-based mapping project in Dar Es Salaam, Tanzania. Dar Ramani Huria (which is Swahili for “Dar Open Map”) is training teams of local university students and community members from throughout Dar Es Salaam to use OpenStreetMap to create sophisticated and highly-accurate maps of Dar es Salaam. These neighborhoods (known as wards) were selected because the are the most flood-prone areas of the city." http://ramanihuria.org/about/
And the Mapillary blog post:
- http://blog.mapillary.com/update/2015/12/17/worldbank.html
And the Humanitarian OSM blogpost:
"A day mapping in Dar Es Salaam: drainage in Chang'ombe"
https://hotosm.org/updates/2016-02-16_a_day_mapping_in_dar_es_salaam_d…