SemiStarGAN: Semi-supervised Generative Adversarial Networks for Multi-domain Image-to-Image Translation

Shu-Yu Hsu, Chih-Yuan Yang, Chi-Chia Huang, Jane Yung-jen Hsu

Abstract: Recent studies have shown significant advance for multi-domain image-to-image translation, and generative adversarial networks (GANs) are widely used to address this problem. However, to train an effective image generator, existing methods all require a large number of domain-labeled images, which may take time and effort to collect for real-world problems. In this paper, we propose SemiStarGAN, a semi-supervised GAN network to tackle this issue. The proposed method utilizes unlabeled images by incorporating a novel discriminator/classifier network architecture—Y model, and two existing semi-supervised learning techniques—pseudo labeling and self-ensembling. Experimental results on the CelebA dataset using domains of facial attributes show that the proposed method achieves comparable performance with state-of-the-art methods using considerably less labeled training images.