Skip to main navigation Skip to search Skip to main content

Conditional image generation using feature-matching GAN

  • Yuzhong Liu
  • , Qiyang Zhao*
  • , Cheng Jiang
  • *Corresponding author for this work
  • Beihang University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Generative Adversarial Net is a frontier method of generative models for images, audios and videos. In this paper, we focus on conditional image generation and introduce conditional Feature-Matching Generative Adversarial Net to generate images from category labels. By visualizing state-of-art discriminative conditional generative models, we find these networks do not gain clear semantic concepts. Thus we design the loss function in the light of metric learning to measure semantic distance. The proposed model is evaluated on several well-known datasets. It is shown to be of higher perceptual quality and better diversity then existing generative models.

Original languageEnglish
Title of host publicationProceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017
EditorsQingli Li, Lipo Wang, Mei Zhou, Li Sun, Song Qiu, Hongying Liu
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-5
Number of pages5
ISBN (Electronic)9781538619377
DOIs
StatePublished - 2 Jul 2017
Event10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017 - Shanghai, China
Duration: 14 Oct 201716 Oct 2017

Publication series

NameProceedings - 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017
Volume2018-January

Conference

Conference10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2017
Country/TerritoryChina
CityShanghai
Period14/10/1716/10/17

Keywords

  • Deep Generative Model
  • Generative Adversarial Net
  • Image Generation

Fingerprint

Dive into the research topics of 'Conditional image generation using feature-matching GAN'. Together they form a unique fingerprint.

Cite this