Face verification using convolutional neural networks with Siamese architecture
Zuzana Bukovčiková, Dominik Sopiak, Miloš Oravec, Jarmila Pavlovičová · 2017
This paper evaluates the ability of convolutional networks to solve the problems arising with face classification in unconstrained environment. It contains design and implementation of Siamese architecture consisting of two convolutional networks used for face verification on sets of photographs. In the scope of the paper, training process is closely monitored and we evaluate several practices and parameters as well as their impact on the network learning.