An Assessment of GANs for Identity-related Applications
This work addresses the problem of assessing GANs' ability to generate new identities, which is crucial for applications like anonymization and data augmentation in identity-related fields.
This paper assesses the ability of GANs to generate new identities for identity-related applications. It concludes that GANs can generate new, imagined identities, making them viable for anonymization and training dataset augmentation. The paper also proposes a novel GAN triplet loss that improves the disentanglement of identity from other image characteristics.
Generative Adversarial Networks (GANs) are now capable of producing synthetic face images of exceptionally high visual quality. In parallel to the development of GANs themselves, efforts have been made to develop metrics to objectively assess the characteristics of the synthetic images, mainly focusing on visual quality and the variety of images. Little work has been done, however, to assess overfitting of GANs and their ability to generate new identities. In this paper we apply a state of the art biometric network to various datasets of synthetic images and perform a thorough assessment of their identity-related characteristics. We conclude that GANs can indeed be used to generate new, imagined identities meaning that applications such as anonymisation of image sets and augmentation of training datasets with distractor images are viable applications. We also assess the ability of GANs to disentangle identity from other image characteristics and propose a novel GAN triplet loss that we show to improve this disentanglement.