Turn a Tesla into a mapping vehicle with Mapillary

Tesla vehicles have often been described as “computers on wheels”. They also include cameras for assisted driving features. In the Mapillary community, these features have raised the question whether the imagery from the car’s cameras could be collected and uploaded to Mapillary. In this post, we present a solution to do exactly that.

Now I just need a Tesla :slight_smile:

1 Like

Mèh… I was hoping they found a way to actually stitch those 4 images together :stuck_out_tongue:

Does anyone own a Tesla around here?

I used this example set of (only…) three images:

and created this “panorama”. When I read documentation of the Tesla correctly there should also be a rear view camera? And also newer models have 8 (!!) cameras… which should make creating an better panorama easier…


I couldn’t find camera specs so I guessed them, but visually the effect is acceptable, I think this is better than the three separate images they have now, take a look:

(Stitching done with Hugin)

[update]
I found the “rear view”. This has quite a different lens… without knowing the lens parameters it would take much more time to figure out the right settings… with this added image, I couln’t get a “fast acceptable result…”

https://www.geoarchief.nl/test/test.html?tesla3

\ther big problem is that the camera to the front and at the sides are relatively in the same position but the camera out the back is probably about 2 to 3 meters “behind”… tried to use the next image in the sequence… but that only made things worse… without more information and preferably more cameras (newer tesla?) I’dd suggest not use the rear camera and use that one as a separate dataset…

HQ | Blog | View / Upload Photos | Become a Trekker | Facebook | Instagram