Most visual servoing schemes are based on the learning of a reference image with the same camera used for servoing. The scheme proposed in this paper differs from previous ones since it is independent on the camera used for learning. With the new scheme it is possible to position a camera (with eventually varying intrinsic parameters) with respect to a non-planar object given a reference image taken with a completely different camera. This promising approach has been successfully tested with an eye-in-hand robotic system.