Strange othophoto result

Hi, I tested WebODM on a dataset of 95 images acquired in a field using a UGV and I got a very strange result :


here is what it should look like (obtained using image warping after estimating homographies with OpenCV):

Is there an explanation for this surprising output? Is there a solution to solve the issue?
It would be helpful also just to know if you encountered similar outputs and in what context. Thanks in advance,

1 Like

Usually when I have seen similar results it has been due to lack of overlap and side lap. What overlap and side lap was this flown with?

Thank you for your reply. These images were taken along a single line (with a ground vehicule) so there is no side lap. The overlap is around 70%. Does the acquisition along a single line make the reconstruction impossible?

I wouldn’t say impossible, but highly unlikely, as it violates some assumptions that OpenDroneMap is built on.

Homographies are the way to go for what you need, not a full photogrammetry tool.

3 Likes

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.