Learning Deformable Models

It is widely recognized that the fundamental building block in high level computer vision is the deformable template, which represents realizations of an object class in the image as noisy geometric i