Resampling using pre-specified points

[nmr2936]

Hi Daniel,

We're new to CloudCompare (CC), and we were wondering if CC can do the following:

We have a bunch of 3D models of human feet obtained using a 3D scanner. The feet have different number of vertices (10000 to 12000, depending on the foot) and, obviously, different number of faces. Also, the vertices do NOT correspond across feet. For instance, vertex 3729 might be on the big toe in one model, on the pinky toe in another model, and near the heel in yet another model.

We'd need to create a dataset of feet that have two characteristics.
(1) All models have the same number of vertices, N.
(2) The vertices must correspond exactly from one foot to the next. For example, if we pick a some vertex Vi at the tip of the pinky toe on one foot, that vertex be at the tip of the pinky toe across all feet.

So-- we think that re-sampling all feet models using the same set of N points will accomplish both of the above goals.

Our question to you is: Is this possible resample pointclouds in CC using a user-specified set of points?

Thank you very much in advance!
Jay

[daniel]

Looks like someone is trying to create a statistical atlas ;)

And no, CloudCompare can't do this. Maybe Meshlab can?

[nmr2936]

Hi Daniel,

Thank you very much for letting me know -- and so promptly, no less! Yes indeed, we're trying to create an atlas.

Following your suggestion, I posted the same question to MeshLab, and I haven't gotten any responses yet. (I had figured that, if I got a prompt response, I'd let you know so that you can pass on that info if anyone else asks you again, but no luck so far.)

Anyway, thank you very much once again,
Jay

[daniel]

Have you looked at Statismo? https://github.com/statismo/statismo

[Alex Carter-Jones]

Hi Jay,

I am currently in search of a database of 3D point cloud data of feet for my biomedical engineering master's project and would be very grateful if you would share your dataset with me?

I can confirm it will not be used for commercial reasons.

Many thanks,

Alex