TY - GEN
T1 - MMGAN
T2 - 24th International Conference on Pattern Recognition, ICPR 2018
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
N1 - Funding Information:
This work was supported by the National Research Council of Science & Technology (NST) grant by the Korea government (MSIP) [No. CRC-15-05-ETRI].
Publisher Copyright:
© 2018 IEEE.
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
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.
Y2 - 20 August 2018 through 24 August 2018
ER -