Pose tags reading issue

[smathermather]

How did you install ODM? (Docker, installer, natively, ...)?

WebODM installer on Ubuntu (Docker)

What is the problem?

Pose (pitch, roll, yaw -- sans GNSS) tags are a nightmare, but found an interesting case here:
https://community.opendronemap.org/t/messy-point-clouds/22635

where if images are resized in WebODM, correct tags are picked up in the ODM engine.

[Type answer here]

What should be the expected behavior? If this is a feature request, please describe in detail the changes you think should be made to the code, citing files and lines where changes should be made, if possible.

Ideally, we read tags when we can and use them to initiate the pose graph.

How can we reproduce this? What steps did you do to trigger the problem? If this is an issue with processing a dataset, YOU MUST include a copy of your dataset AND task output log, uploaded on Google Drive or Dropbox (otherwise we cannot reproduce this).

Here is the full dataset (including procedd products if resize isn't performed, and thus tags corrected:
https://crankyserver.com/public/task/7ed888b8-fe8d-4bd1-82da-a98baa8554fc/map/?t=orthophoto

And here it is with resizing with a sensible output:
https://crankyserver.com/public/task/9ef13df4-e6f0-4e10-8ad5-285afb6e5248/map/?t=orthophoto

BTW, one parameter change -- zipping was taking forever with the resized one, so I did add auto-boundary and auto-boundary distance just to get the data clipped to a reasonable extent.

[smathermather]

It seems perhaps additional work on DJI's part to change where they are storing tags, this one burying GNSS tags, and ODM not picking it up correctly even with resizing:

https://crankyserver.com/public/task/291f2e3e-a680-4bee-98a9-e145423fdc82/map/?t=orthophoto