It has ben discussed a few times before I’d like to push it even further.
I’d like the Align operation to be a completely separate function from merging.
Why?
It is useful to do some edits after alignment but before merging.
When we’re aligning two clouds it’s better to have as much geometry in both clouds as possible. Even if it’s not perfect.
On a 1m object you’ll get better alignment from two clouds, both having some detail (even rough) from the entire 1m space of the object than, from a perfect overlap of just 10cm.
So when we’re aligning two clouds we want as much data in both clouds as possible to get the best alignment possible.
Then when the two clouds are aligned, we can see how they align to each other and decide which part of the model we want from cloud1 and which parts we want from cloud2.
At this point we should be able to view both clouds on top of each other and at the same time be able to select fragments of one cloud and delete them - this way shaping the clouds to perfectly complete each other.
After that we can merge the two clouds together.
It would be even better if we could align various clouds first - there should be a clear indication of which cloud is parent and which cloud is being aligned to the parent - this way we could align more then two clouds together.
Then we could edit those clouds and finally merge several clouds together as a single final operation.
That is a workflow from other point cloud software as well as from photogrammetry software and it’s really missing from Revoscan 