Idiap - Resnet V2/V1 - Casia Webface

Inspired by **FaceNet** we here at Idiap trained our own CNN using the Inception Resnet 2 architecture using Casia Webface database. In this links you can find the script that trains this neural network.

To trigger this training it’s necessary to use the bob.learn.tensorflow package and run the following command:

$ ./bin/jman submit --name CELEB-GRAY --queue gpu -- bob_tf_train_generic CASIA_inception_resnet_v2_center_loss.py

Some quick details about this CNN (just as a mental note):

• The hot encoded layer has 10574 neurons.

• Faces were detected and croped to \(182 \times 182\) using MTCNN face and landmark detector

• The following data augmentation strategies were implemented:

  • Random crop to \(160 \times 160\)

  • Random Flip

  • Images were normalized to have zero mean and standard deviation one

• Learning rate of 0.1, 0.01, 0.001

• RMSPROP Optimizer

• Batch size of 90