Unpaired Multi-Domain Image Generation via Regularized Conditional GANs

Research output: Chapters, Conference Papers, Creative and Literary Works (RGC: 12, 32, 41, 45)32_Refereed conference paper (with ISBN/ISSN)

View graph of relations

Related Research Unit(s)

Detail(s)

Original languageEnglish
Title of host publicationProceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI-18)
EditorsJérôme Lang
PublisherInternational Joint Conferences on Artificial Intelligence
Pages2553-2559
ISBN (Electronic)9780999241127
StateAccepted/In press/Filed - Jul 2018

Conference

Title27th International Joint Conference on Artificial Intelligence and the 23rd European Conference on Artificial Intelligence (IJCAI-ECAI 2018)
LocationStockholmsmässan – The Stockholm Convention Center
PlaceSweden
CityStockholm
Period13 - 19 July 2018

Abstract

In this paper, we study the problem of multi-domain image generation, the goal of which is to generate pairs of corresponding images from different domains. With the recent development in generative models, image generation has achieved great progress and has been applied to various computer vision tasks. However, multi-domain image generation may not achieve the desired performance due to the difficulty of learning the correspondence of different domain images, especially when the information of paired samples is not given. To tackle this problem, we propose Regularized Conditional GAN (RegCGAN) which is capable of learning to generate corresponding images in the absence of paired training data. RegCGAN is based on the conditional GAN, and we introduce two regularizers to guide the model to learn the corresponding semantics of different domains. We evaluate the proposed model on several tasks for which paired training data is not given, including the generation of edges and photos, the generation of faces with different attributes, etc. The experimental results show that our model can successfully generate corresponding images for all these tasks, while outperforms the baseline methods. We also introduce an approach of applying RegCGAN to unsupervised domain adaptation.

Citation Format(s)

Unpaired Multi-Domain Image Generation via Regularized Conditional GANs. / Mao, Xudong; Li, Qing.

Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI-18). ed. / Jérôme Lang. International Joint Conferences on Artificial Intelligence, 2018. p. 2553-2559.

Research output: Chapters, Conference Papers, Creative and Literary Works (RGC: 12, 32, 41, 45)32_Refereed conference paper (with ISBN/ISSN)