DisguiseNet : A Contrastive Approach for Disguised Face Verification in the Wild

This paper describes our approach for the Disguised Faces in the Wild (DFW)2018 challenge. The task here is to verify the identity of a person amongdisguised and impostors images. Given the importance of the task of faceverification it is essential to compare methods across a common platform. Ourapproach is based on VGG-face architecture paired with Contrastive loss basedon cosine distance metric. For augmenting the data set, we source more datafrom the internet. The experiments show the effectiveness of the approach onthe DFW data. We show that adding extra data to the DFW dataset with noisylabels also helps in increasing the generalization performance of the network.The proposed network achieves 27.13% absolute increase in accuracy over the DFWbaseline.