MMGAN: Manifold-Matching Generative Adversarial Networks

Noseong Park, Ankesh Anand, Joel Ruben Antony Moniz, Kookjin Lee, Jaegul Choo, David Keetae Park, Tanmoy Chakraborty, Hongkyu Park, Youngmin Kim

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

Abstract

It is well-known that GANs are difficult to train, and several different techniques have been proposed in order to stabilize their training. In this paper, we propose a novel training method called manifold-matching, and a new GAN model called manifold-matching GAN (MMGAN). MMGAN finds two manifolds representing the vector representations of real and fake images. If these two manifolds match, it means that real and fake images are statistically identical. To assist the manifold-matching task, we also use i) kernel tricks to find better manifold structures, ii) moving-averaged manifolds across mini-batches, and iii) a regularizer based on correlation matrix to suppress mode collapse. We conduct in-depth experiments with three image datasets and compare with several state-of-the-art GAN models. 32.4% of images generated by the proposed MMGAN are recognized as fake images during our user study (16% enhancement compared to other state-of-the-art model). MMGAN achieved an unsupervised inception score of 7.8 for CIFAR-10.

Original languageEnglish
Title of host publication2018 24th International Conference on Pattern Recognition, ICPR 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1343-1348
Number of pages6
ISBN (Electronic)9781538637883
DOIs
Publication statusPublished - 2018 Nov 26
Event24th International Conference on Pattern Recognition, ICPR 2018 - Beijing, China
Duration: 2018 Aug 202018 Aug 24

Publication series

NameProceedings - International Conference on Pattern Recognition
Volume2018-August
ISSN (Print)1051-4651

Conference

Conference24th International Conference on Pattern Recognition, ICPR 2018
CountryChina
CityBeijing
Period18/8/2018/8/24

Fingerprint

Experiments

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Park, N., Anand, A., Moniz, J. R. A., Lee, K., Choo, J., Park, D. K., ... Kim, Y. (2018). MMGAN: Manifold-Matching Generative Adversarial Networks. In 2018 24th International Conference on Pattern Recognition, ICPR 2018 (pp. 1343-1348). [8545881] (Proceedings - International Conference on Pattern Recognition; Vol. 2018-August). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICPR.2018.8545881

MMGAN : Manifold-Matching Generative Adversarial Networks. / Park, Noseong; Anand, Ankesh; Moniz, Joel Ruben Antony; Lee, Kookjin; Choo, Jaegul; Park, David Keetae; Chakraborty, Tanmoy; Park, Hongkyu; Kim, Youngmin.

2018 24th International Conference on Pattern Recognition, ICPR 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 1343-1348 8545881 (Proceedings - International Conference on Pattern Recognition; Vol. 2018-August).

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

Park, N, Anand, A, Moniz, JRA, Lee, K, Choo, J, Park, DK, Chakraborty, T, Park, H & Kim, Y 2018, MMGAN: Manifold-Matching Generative Adversarial Networks. in 2018 24th International Conference on Pattern Recognition, ICPR 2018., 8545881, Proceedings - International Conference on Pattern Recognition, vol. 2018-August, Institute of Electrical and Electronics Engineers Inc., pp. 1343-1348, 24th International Conference on Pattern Recognition, ICPR 2018, Beijing, China, 18/8/20. https://doi.org/10.1109/ICPR.2018.8545881
Park N, Anand A, Moniz JRA, Lee K, Choo J, Park DK et al. MMGAN: Manifold-Matching Generative Adversarial Networks. In 2018 24th International Conference on Pattern Recognition, ICPR 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 1343-1348. 8545881. (Proceedings - International Conference on Pattern Recognition). https://doi.org/10.1109/ICPR.2018.8545881
Park, Noseong ; Anand, Ankesh ; Moniz, Joel Ruben Antony ; Lee, Kookjin ; Choo, Jaegul ; Park, David Keetae ; Chakraborty, Tanmoy ; Park, Hongkyu ; Kim, Youngmin. / MMGAN : Manifold-Matching Generative Adversarial Networks. 2018 24th International Conference on Pattern Recognition, ICPR 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 1343-1348 (Proceedings - International Conference on Pattern Recognition).
@inproceedings{d5925315c05644f18db349296ef8672f,
title = "MMGAN: Manifold-Matching Generative Adversarial Networks",
abstract = "It is well-known that GANs are difficult to train, and several different techniques have been proposed in order to stabilize their training. In this paper, we propose a novel training method called manifold-matching, and a new GAN model called manifold-matching GAN (MMGAN). MMGAN finds two manifolds representing the vector representations of real and fake images. If these two manifolds match, it means that real and fake images are statistically identical. To assist the manifold-matching task, we also use i) kernel tricks to find better manifold structures, ii) moving-averaged manifolds across mini-batches, and iii) a regularizer based on correlation matrix to suppress mode collapse. We conduct in-depth experiments with three image datasets and compare with several state-of-the-art GAN models. 32.4{\%} of images generated by the proposed MMGAN are recognized as fake images during our user study (16{\%} enhancement compared to other state-of-the-art model). MMGAN achieved an unsupervised inception score of 7.8 for CIFAR-10.",
author = "Noseong Park and Ankesh Anand and Moniz, {Joel Ruben Antony} and Kookjin Lee and Jaegul Choo and Park, {David Keetae} and Tanmoy Chakraborty and Hongkyu Park and Youngmin Kim",
year = "2018",
month = "11",
day = "26",
doi = "10.1109/ICPR.2018.8545881",
language = "English",
series = "Proceedings - International Conference on Pattern Recognition",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "1343--1348",
booktitle = "2018 24th International Conference on Pattern Recognition, ICPR 2018",

}

TY - GEN

T1 - MMGAN

T2 - Manifold-Matching Generative Adversarial Networks

AU - Park, Noseong

AU - Anand, Ankesh

AU - Moniz, Joel Ruben Antony

AU - Lee, Kookjin

AU - Choo, Jaegul

AU - Park, David Keetae

AU - Chakraborty, Tanmoy

AU - Park, Hongkyu

AU - Kim, Youngmin

PY - 2018/11/26

Y1 - 2018/11/26

N2 - It is well-known that GANs are difficult to train, and several different techniques have been proposed in order to stabilize their training. In this paper, we propose a novel training method called manifold-matching, and a new GAN model called manifold-matching GAN (MMGAN). MMGAN finds two manifolds representing the vector representations of real and fake images. If these two manifolds match, it means that real and fake images are statistically identical. To assist the manifold-matching task, we also use i) kernel tricks to find better manifold structures, ii) moving-averaged manifolds across mini-batches, and iii) a regularizer based on correlation matrix to suppress mode collapse. We conduct in-depth experiments with three image datasets and compare with several state-of-the-art GAN models. 32.4% of images generated by the proposed MMGAN are recognized as fake images during our user study (16% enhancement compared to other state-of-the-art model). MMGAN achieved an unsupervised inception score of 7.8 for CIFAR-10.

AB - It is well-known that GANs are difficult to train, and several different techniques have been proposed in order to stabilize their training. In this paper, we propose a novel training method called manifold-matching, and a new GAN model called manifold-matching GAN (MMGAN). MMGAN finds two manifolds representing the vector representations of real and fake images. If these two manifolds match, it means that real and fake images are statistically identical. To assist the manifold-matching task, we also use i) kernel tricks to find better manifold structures, ii) moving-averaged manifolds across mini-batches, and iii) a regularizer based on correlation matrix to suppress mode collapse. We conduct in-depth experiments with three image datasets and compare with several state-of-the-art GAN models. 32.4% of images generated by the proposed MMGAN are recognized as fake images during our user study (16% enhancement compared to other state-of-the-art model). MMGAN achieved an unsupervised inception score of 7.8 for CIFAR-10.

UR - http://www.scopus.com/inward/record.url?scp=85059743190&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059743190&partnerID=8YFLogxK

U2 - 10.1109/ICPR.2018.8545881

DO - 10.1109/ICPR.2018.8545881

M3 - Conference contribution

AN - SCOPUS:85059743190

T3 - Proceedings - International Conference on Pattern Recognition

SP - 1343

EP - 1348

BT - 2018 24th International Conference on Pattern Recognition, ICPR 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -