Transfer Learning for Facial Recognition

transfer_learning.ipynb:

Repurposed the convolutional base from InceptionV3 (with pretrained weights from ImageNet), to differentiate between human faces. Added 2 dense layers on top of the convolutional base and trained the new layers on 38 human faces.

CroppedYale/ :

Folder with data as received from Yale Face Database.

data/ :

Reformatted data to work smoothly with Keras interface.