Hi Alex,
Yes, effectively to merge two or more shapes into a single shape. Sorry for not being clear.
To give an example:
We detect (independently) objects within the scene and know the location on individual photos. Since the photos have overlapping views of the scene, the same object is detected multiple times. The object detection part of the code is oblivious to this. Hence at the time of mapping the polygons to the model, we end up with overlapping (but slightly different) shapes depicting the same physical entity in the scene.
If there is some built-in functionality (or a particular way of dealing with it) in metashape to handle this, that is what I am after.
Regards,
Raj.