Propose a version of the subspace identity model that is robust to outliers in the training data.
Moghaddam et al. (2000) took a different probabilistic approach to face verification. Theytookthedifferencex∆=x2−x1andmodeledthelikelihoodsofthisvectorPr(x∆|w=0)andPr(x∆|w=1)whenthetwofacesmatchordon’t.Proposeexpressionsfortheselikelihoodsanddiscuss learning and inference in this model. Identify one possible disadvantage of this model.
Already registered? Login
Not Account? Sign up
Enter your email address to reset your password
Back to Login? Click here