Create bbox around point cloud by camera names?

[photoscan_user]

Is there any script to create bbox around point cloud by camera names?

[Alexey Pasumansky]

Can you please elaborate this in additional details?

[photoscan_user]

For example I have some large chunk for which I have created sparse point cloud and then I want to compute bounding box for some part of this chunk based on image names, this bounding box will be used for other steps of pipeline like dense cloud generation, etc.

[Alexey Pasumansky]

Hello photoscan_user,

What would be passed to the script input -the pairs of camera labels? Should them be used for the bounding box size definition only or for the orientation as well and the bounding box center?

[photoscan_user]

Input should be list of image names (camera names) which is subset of images of large chunk.

Photoscan already have bbox for large chunk, so I guess orientation of computed bbox should be the same by Z coordinate, because it's important for orthomap generation. I don't know about the center of bbox, it's needed only if it is used by Photoscan somewhere under the hood.

[Samuel]

It would also be helpful to generate the bounding box based on given gps coordinates or have a way to fit the bounding box around the cameras such that nothing (or very little) behind the cameras is processed