ST-GAN: Unsupervised Facial Image Semantic Transformation Using Generative Adversarial Networks

Jichao Zhang, Fan Zhong, Gongze Cao, Xueying Qin
Proceedings of the Ninth Asian Conference on Machine Learning, PMLR 77:248-263, 2017.

Abstract

Image semantic transformation aims to convert one image into another image with different semantic features (e.g., face pose, hairstyle). The previous methods, which learn the mapping function from one image domain to the other, require supervised information directly or indirectly. In this paper, we propose an unsupervised image semantic transformation method called semantic transformation generative adversarial networks (ST-GAN), and experimentally verify it on face dataset. We further improve ST-GAN with the Wasserstein distance to generate more realistic images and propose a method called local mutual information maximization to obtain a more explicit semantic transformation. ST-GAN has the ability to map the image semantic features into the latent vector and then perform transformation by controlling the latent vector.

Cite this Paper

BibTeX
@InProceedings{pmlr-v77-zhang17c, title = {ST-GAN: Unsupervised Facial Image Semantic Transformation Using Generative Adversarial Networks}, author = {Zhang, Jichao and Zhong, Fan and Cao, Gongze and Qin, Xueying}, booktitle = {Proceedings of the Ninth Asian Conference on Machine Learning}, pages = {248--263}, year = {2017}, editor = {Zhang, Min-Ling and Noh, Yung-Kyun}, volume = {77}, series = {Proceedings of Machine Learning Research}, address = {Yonsei University, Seoul, Republic of Korea}, month = {15--17 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v77/zhang17c/zhang17c.pdf}, url = {https://proceedings.mlr.press/v77/zhang17c.html}, abstract = {Image semantic transformation aims to convert one image into another image with different semantic features (e.g., face pose, hairstyle). The previous methods, which learn the mapping function from one image domain to the other, require supervised information directly or indirectly. In this paper, we propose an unsupervised image semantic transformation method called semantic transformation generative adversarial networks (ST-GAN), and experimentally verify it on face dataset. We further improve ST-GAN with the Wasserstein distance to generate more realistic images and propose a method called local mutual information maximization to obtain a more explicit semantic transformation. ST-GAN has the ability to map the image semantic features into the latent vector and then perform transformation by controlling the latent vector.} }
Endnote
%0 Conference Paper %T ST-GAN: Unsupervised Facial Image Semantic Transformation Using Generative Adversarial Networks %A Jichao Zhang %A Fan Zhong %A Gongze Cao %A Xueying Qin %B Proceedings of the Ninth Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2017 %E Min-Ling Zhang %E Yung-Kyun Noh %F pmlr-v77-zhang17c %I PMLR %P 248--263 %U https://proceedings.mlr.press/v77/zhang17c.html %V 77 %X Image semantic transformation aims to convert one image into another image with different semantic features (e.g., face pose, hairstyle). The previous methods, which learn the mapping function from one image domain to the other, require supervised information directly or indirectly. In this paper, we propose an unsupervised image semantic transformation method called semantic transformation generative adversarial networks (ST-GAN), and experimentally verify it on face dataset. We further improve ST-GAN with the Wasserstein distance to generate more realistic images and propose a method called local mutual information maximization to obtain a more explicit semantic transformation. ST-GAN has the ability to map the image semantic features into the latent vector and then perform transformation by controlling the latent vector.
APA
Zhang, J., Zhong, F., Cao, G. & Qin, X.. (2017). ST-GAN: Unsupervised Facial Image Semantic Transformation Using Generative Adversarial Networks. Proceedings of the Ninth Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 77:248-263 Available from https://proceedings.mlr.press/v77/zhang17c.html.

Related Material