Font creation using class discriminative deep convolutional generative adversarial networks

Kotaro Abe, Brian Kenji Iwana, Viktor Gosta Holmer, Seiichi Uchida

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

Abstract

In this research, we attempt to generate fonts automatically using a modification of a Deep Convolutional Generative Adversarial Network (DCGAN) by introducing class consideration. DCGANs are the application of generative adversarial networks (GAN) which make use of convolutional and deconvolutional layers to generate data through adversarial detection. The conventional GAN is comprised of two neural networks that work in series. Specifically, it approaches an unsupervised method of data generation with the use of a generative network whose output is fed into a second discriminative network. While DCGANs have been successful on natural images, we show its limited ability on font generation due to the high variation of fonts combined with the need of rigid structures of characters. We propose a class discriminative DCGAN which uses a classification network to work alongside the discriminative network to refine the generative network. This results of our experiment shows a dramatic improvement over the conventional DCGAN.

Original languageEnglish
Title of host publicationProceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages238-243
Number of pages6
ISBN (Electronic)9781538633540
DOIs
Publication statusPublished - Dec 13 2018
Event4th Asian Conference on Pattern Recognition, ACPR 2017 - Nanjing, China
Duration: Nov 26 2017Nov 29 2017

Publication series

NameProceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017

Other

Other4th Asian Conference on Pattern Recognition, ACPR 2017
CountryChina
CityNanjing
Period11/26/1711/29/17

Fingerprint

Rigid structures
Neural networks
Experiments

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Abe, K., Iwana, B. K., Holmer, V. G., & Uchida, S. (2018). Font creation using class discriminative deep convolutional generative adversarial networks. In Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017 (pp. 238-243). [8575829] (Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ACPR.2017.99

Font creation using class discriminative deep convolutional generative adversarial networks. / Abe, Kotaro; Iwana, Brian Kenji; Holmer, Viktor Gosta; Uchida, Seiichi.

Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017. Institute of Electrical and Electronics Engineers Inc., 2018. p. 238-243 8575829 (Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017).

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

Abe, K, Iwana, BK, Holmer, VG & Uchida, S 2018, Font creation using class discriminative deep convolutional generative adversarial networks. in Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017., 8575829, Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017, Institute of Electrical and Electronics Engineers Inc., pp. 238-243, 4th Asian Conference on Pattern Recognition, ACPR 2017, Nanjing, China, 11/26/17. https://doi.org/10.1109/ACPR.2017.99
Abe K, Iwana BK, Holmer VG, Uchida S. Font creation using class discriminative deep convolutional generative adversarial networks. In Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017. Institute of Electrical and Electronics Engineers Inc. 2018. p. 238-243. 8575829. (Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017). https://doi.org/10.1109/ACPR.2017.99
Abe, Kotaro ; Iwana, Brian Kenji ; Holmer, Viktor Gosta ; Uchida, Seiichi. / Font creation using class discriminative deep convolutional generative adversarial networks. Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 238-243 (Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017).
@inproceedings{6f301bd9027843069fc3ad3c04d5822c,
title = "Font creation using class discriminative deep convolutional generative adversarial networks",
abstract = "In this research, we attempt to generate fonts automatically using a modification of a Deep Convolutional Generative Adversarial Network (DCGAN) by introducing class consideration. DCGANs are the application of generative adversarial networks (GAN) which make use of convolutional and deconvolutional layers to generate data through adversarial detection. The conventional GAN is comprised of two neural networks that work in series. Specifically, it approaches an unsupervised method of data generation with the use of a generative network whose output is fed into a second discriminative network. While DCGANs have been successful on natural images, we show its limited ability on font generation due to the high variation of fonts combined with the need of rigid structures of characters. We propose a class discriminative DCGAN which uses a classification network to work alongside the discriminative network to refine the generative network. This results of our experiment shows a dramatic improvement over the conventional DCGAN.",
author = "Kotaro Abe and Iwana, {Brian Kenji} and Holmer, {Viktor Gosta} and Seiichi Uchida",
year = "2018",
month = "12",
day = "13",
doi = "10.1109/ACPR.2017.99",
language = "English",
series = "Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "238--243",
booktitle = "Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017",
address = "United States",

}

TY - GEN

T1 - Font creation using class discriminative deep convolutional generative adversarial networks

AU - Abe, Kotaro

AU - Iwana, Brian Kenji

AU - Holmer, Viktor Gosta

AU - Uchida, Seiichi

PY - 2018/12/13

Y1 - 2018/12/13

N2 - In this research, we attempt to generate fonts automatically using a modification of a Deep Convolutional Generative Adversarial Network (DCGAN) by introducing class consideration. DCGANs are the application of generative adversarial networks (GAN) which make use of convolutional and deconvolutional layers to generate data through adversarial detection. The conventional GAN is comprised of two neural networks that work in series. Specifically, it approaches an unsupervised method of data generation with the use of a generative network whose output is fed into a second discriminative network. While DCGANs have been successful on natural images, we show its limited ability on font generation due to the high variation of fonts combined with the need of rigid structures of characters. We propose a class discriminative DCGAN which uses a classification network to work alongside the discriminative network to refine the generative network. This results of our experiment shows a dramatic improvement over the conventional DCGAN.

AB - In this research, we attempt to generate fonts automatically using a modification of a Deep Convolutional Generative Adversarial Network (DCGAN) by introducing class consideration. DCGANs are the application of generative adversarial networks (GAN) which make use of convolutional and deconvolutional layers to generate data through adversarial detection. The conventional GAN is comprised of two neural networks that work in series. Specifically, it approaches an unsupervised method of data generation with the use of a generative network whose output is fed into a second discriminative network. While DCGANs have been successful on natural images, we show its limited ability on font generation due to the high variation of fonts combined with the need of rigid structures of characters. We propose a class discriminative DCGAN which uses a classification network to work alongside the discriminative network to refine the generative network. This results of our experiment shows a dramatic improvement over the conventional DCGAN.

UR - http://www.scopus.com/inward/record.url?scp=85060513937&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060513937&partnerID=8YFLogxK

U2 - 10.1109/ACPR.2017.99

DO - 10.1109/ACPR.2017.99

M3 - Conference contribution

AN - SCOPUS:85060513937

T3 - Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017

SP - 238

EP - 243

BT - Proceedings - 4th Asian Conference on Pattern Recognition, ACPR 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -