Hi, I'm trying to figure out the logic behind the robust nonrigid alignment examples. Could you refer me to the paper about it? Thanks in advance.