High-Resolution Deep Convolutional Generative Adversarial Networks
release_iq4qzqx6cfamxpsep4wnh36y7a
by
J. D. Curtó and H. C. Zarza and T. Kim
2018
Abstract
Generative Adversarial Networks (GANs) [Goodfellow et al. 2014] convergence
in a high-resolution setting with a computational constrain of GPU memory
capacity has been beset with difficulty due to the known lack of convergence
rate stability. In order to boost network convergence of DCGAN (Deep
Convolutional Generative Adversarial Networks) [Radford et al. 2016] and
achieve good-looking high-resolution results we propose a new layered network
structure, HDCGAN, that incorporates current state-of-the-art techniques for
this effect. Glasses, a mechanism to arbitrarily improve the final GAN
generated results by enlarging the input size by a telescope ζ is also
presented. A novel bias-free dataset, Graphics, containing human faces from
different ethnical groups in a wide variety of illumination conditions and
image resolutions is introduced. Graphics is enhanced with HDCGAN synthetic
images, thus being the first GAN augmented face dataset. We conduct extensive
experiments on CelebA [Liu et al. 2015], CelebA-hq [Karras et al. 2018] and
Graphics. HDCGAN is the current state-of-the-art in synthetic image generation
on CelebA achieving a MS-SSIM of 0.1978 and a FRÉCHET Inception Distance of
8.44.
In text/plain
format
Archived Files and Locations
application/pdf 10.2 MB
file_qxjcyoiw3bh27ivqgnbnky4xai
|
arxiv.org (repository) web.archive.org (webarchive) |
1711.06491v11
access all versions, variants, and formats of this works (eg, pre-prints)