Error report: First image and last image of a sequence connected by a straight line

Hi all

After I uploaded one of my sequences I noticed that the first image and the last image of the sequence are linked and shown with a straight connected line.

Image sequence: sqnbjubn9fqhn3jtwoswtp

Can anyone take a look?

I saw a similar issue reported some time ago, but that was on Android, and I used iOS app. Not sure if they are related.

http://mapillary.trydiscourse.com/t/the-first-point-from-the-sequence-connects-to-the-last-one/1125

Can you check image dates on the “snap point”?
I’dd expect that something went wrong there?
The “snap-point” is somewhere in the middle of the sequence… so that is odd…

I found the Mapillary link:

Sorry, I don’t understand what you mean by ‘snap-point’ and where in the middle of the sequence?

The dates of May 21 2019 is correct for all the images.

With “snap point” I mean the point where one image is at one end of the long line (you don’t want/expect) and the next is at the other end.

The link I provided in the previous point is before the snap or jump. When you click the ‘>’ you jump to the other image. What are the timestamps on those two images, are they wrong? Mapillary ordeers the sequence based upon timestamp. If the timestamps are wrong you get this…

Usually one starts on point A on for example 14:00 and passes point B on on for example 14:30 and ends at point C on for example 15:00.

Sounds obvious maybe :wink:

It looks like your sequence starts at 14:00 on point B, is at point C on 14:30 and at point A at 14:31 and ends at point B at 15:00!?

Take a look at the start of your sequence, it just before Huntingdon… that’s odd isn’t it?

Ah OK. I understand now. So, the time stamp of the very first image in the sequence is 13:24, and then Huntington, the sequence appeared to have been split and the last image of the first part of the sequence ended at 13:59 (image key: qqOZIIqxgiqjJdvNGpLctw) and image at the start of the next sequence (7-wvhdeEL2XEG5cws-vJwA) shows a time stamp of 02:00, rather than 14:00, and all other subsequent images have the same time stamp format.

Could it be a bug after the sequence was split?

(Ah, you found the time!
Do you haven image viewer that can read EXIF data? (Take a look at Irfanview, takes a little getting used to, but’s quite powerfull!)

Hmmm 2:00 and 14:00 smells like a bug somewhere in a conversion error from/to 24 hour system and AM/PM structure?!

Or you could upload a few images (pick the interresting ones) once more into the webuploader and look closely to the order it uploads/adds… that’ll tell you also (don’t publish…)