Add Photos inquiry

[AvionAvionics]

I have read through the tutorials and experimenting. When you add photos under workflow I always seems to fail. I have tried choosing 1-5 and and I have also tried many (like 30). But none seem to get added, That being said one set is always High up and another set is much closer (hopes of getting much more detail with the close and broad area) . Do photos that are added need to be the same distance away? The goal is to recreate broad areas but have the ability to have high res models when zoomed. I have achieved this partially through separate scenes and control points but the process is a bit arduous and was hoping for a more refined way.

[Roberto]

Hi AvionAvionics, the 'add photos' routine uses the same matching algorithm used by Samantha. This means that if the structure from motion pipeline didn't oriented all the pictures at once, it's unlikely they will be added with the 'add photos' dialog automatically. Unfortunately manual control points might be the only options for these cases.
Photos at similar distances might be easier to deal with. If you change the distance a lot, I would suggest to also add photos with intermediate zoom / distance level in order to help the matching algorithm.

[AvionAvionics]

Thanks for the answer Roberto. Stupid question for you if you don't mind answering. If I have a dense point cloud that is already built but an area is a little unpopulated and I add photos into the scene (to help give more detail to the unpopulated area) does Zephyr need to re-render the entire point cloud or can it use the old point cloud as a base and just render the point cloud from the new photos.

[AvionAvionics]

Nvm Roberto it appears I just need to dig into the manual more. For anyone else interested it appears that once the photos are added it will replace the old sparse point cloud with a new one and you can generate a dense point cloud from there.

[Roberto]

It's better if you regenerate a new dense point cloud after adding new photos to the workspace. The existing dense point cloud can still be used, but any subsequent step (mesh, texture) will not use the information coming from the newly added photos.