Iconify: Converting Photographs into Icons

Takuro Karamatsu, Gibran Benitez-Garcia, Keiji Yanai, Seiichi Uchida

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In this paper, we tackle a challenging domain conversion task between photo and icon images. Although icons often originate from real object images (i.e., photographs), severe abstractions and simplifications are applied to generate icon images by professional graphic designers. Moreover, there is no one-to-one correspondence between the two domains, for this reason we cannot use it as the ground-truth for learning a direct conversion function. Since generative adversarial networks (GAN) can undertake the problem of domain conversion without any correspondence, we test CycleGAN and UNIT to generate icons from objects segmented from photo images. Our experiments with several image datasets prove that CycleGAN learns sufficient abstraction and simplification ability to generate icon-like images.

Original languageEnglish
Title of host publicationMMArt-ACM 2020 - Proceedings of the 2020 Joint Workshop on Multimedia Artworks Analysis and Attractiveness Computing in Multimedia
PublisherAssociation for Computing Machinery, Inc
Pages7-12
Number of pages6
ISBN (Electronic)9781450371377
DOIs
Publication statusPublished - Jun 8 2020
Event3rd Joint International Workshop on Multimedia Artworks Analysis and International Workshop on Attractiveness Computing in Multimedia, MMArt-ACM 2020 - Dublin, Ireland
Duration: Jun 8 2020 → …

Publication series

NameMMArt-ACM 2020 - Proceedings of the 2020 Joint Workshop on Multimedia Artworks Analysis and Attractiveness Computing in Multimedia

Conference

Conference3rd Joint International Workshop on Multimedia Artworks Analysis and International Workshop on Attractiveness Computing in Multimedia, MMArt-ACM 2020
CountryIreland
CityDublin
Period6/8/20 → …

All Science Journal Classification (ASJC) codes

  • Computer Graphics and Computer-Aided Design
  • Computer Networks and Communications
  • Computer Science Applications

Fingerprint Dive into the research topics of 'Iconify: Converting Photographs into Icons'. Together they form a unique fingerprint.

  • Cite this

    Karamatsu, T., Benitez-Garcia, G., Yanai, K., & Uchida, S. (2020). Iconify: Converting Photographs into Icons. In MMArt-ACM 2020 - Proceedings of the 2020 Joint Workshop on Multimedia Artworks Analysis and Attractiveness Computing in Multimedia (pp. 7-12). (MMArt-ACM 2020 - Proceedings of the 2020 Joint Workshop on Multimedia Artworks Analysis and Attractiveness Computing in Multimedia). Association for Computing Machinery, Inc. https://doi.org/10.1145/3379173.3393708