Instance Map Based Image Synthesis With a Denoising Generative Adversarial Network

Semantic layout-based image synthesizing, which has benefited from the success of generative adversarial networks (GANs), has received a substantial amount of attention recently. How to enhance the synthesis image equality while maintaining the stochasticity of the GAN remains a challenge. We propos...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access 2018-01, Vol.6, p.33654-33665
Main Authors: Zheng, Ziqiang, Wang, Chao, Yu, Zhibin, Zheng, Haiyong, Zheng, Bing
Format: Article
Language:English
Subjects:
artificial neural networks
Convolution
Gallium nitride
Generative adversarial networks
Generators
Image enhancement
Image processing
Layout
Layouts
Neural networks
Noise reduction
Semantics
Synthesis
Task analysis
Underwater technology
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Semantic layout-based image synthesizing, which has benefited from the success of generative adversarial networks (GANs), has received a substantial amount of attention recently. How to enhance the synthesis image equality while maintaining the stochasticity of the GAN remains a challenge. We propose a novel denoising framework to handle this problem. The generation of overlapping objects is another challenging task when synthesizing images from a semantic layout to a realistic RGB photograph. To overcome this deficiency, we include a one-hot semantic label map to force the generator to pay more attention to the generation of overlapping objects. Furthermore, we improve the loss function of the discriminator by considering the perturbed loss and cascade layer loss to guide the generation process. We applied our methods to the Cityscapes, photo-sketch, day-night, facades, and NYU datasets to demonstrate the image generation ability of our model.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2018.2849108