Open AccessComparative Review of Cross-Domain Generative Adversarial Networks
Author(s) -
Bassel Zeno,
Ilya Kalinovskiy,
Yuri Matveev
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/618/1/012012
Subject(s) - generalization , computer science , generative grammar , metric (unit) , set (abstract data type) , artificial intelligence , translation (biology) , image (mathematics) , domain (mathematical analysis) , adversarial system , star (game theory) , image translation , pattern recognition (psychology) , generative model , machine learning , natural language processing , mathematics , mathematical analysis , biochemistry , operations management , chemistry , messenger rna , economics , gene , programming language
This paper provides the comparative analysis between two recent image-to-image translation models that based on Generative Adversarial Networks. The first one is UNIT which consists of coupled GANs and variational autoencoders (VAEs) with shared-latent space, and the second one is Star-GAN which contains a single GAN model. Given training data from two different domains from dataset CelebA, these two models learn translation task in two directions. The term domain denotes as a set of images sharing the same attribute value. So, the attributes that are prepared: eye glasses, blond hair, beard, smiling and age. Five UNIT models are trained separately, while only one Star-GAN model is trained. For evaluation, we conduct some experiments and provide a quantitative comparison using direct metric GAM (Generative Adversarial Metric) to quantify the ability of generalization and the ability of generating photorealistic photos. The experimental results show the superiority of cross-model UNIT over multi-model StarGAN on generating age and eye glasses attributes, and the equivalent performance to synthesize other attributes.