[Deep Learning] Face Recognition

Face verification & face recognition

• face recognition
  • has a database of K person
  • get an input image
  • output id if the image is any of the k persons
• face verification
  • input image, name
  • output whether the input image is that of the claimed person

---

One shot learning

Larning from one example to recognize the person again.

Learning a 'similarity' function
d(img1, img2) = degree of difference between images.
If d(img1, img2) <= t "same"
If d(img1, img2) > t "different"

{verification}

use softmax to classify the faces. Basically use CNN to make feature map of each faces.