Thanks a lot everybody for all your answers and suggestions!
I thought that maybe I could make some use of microsoft kinect or similar device to make the 3d map of the object (since getting the data from simple photos looks incredibly cpu intensive) using some open source elaboration program like PCL ( http://pointclouds.org/ )

Still I'll be extremely interested if anybody could link me some info about 3d objects handling in ruby or would simply have some insight to share on the argument.

Many thanks,
Nicol