Generative Adversarial Image Super-Resolution Through Deep Dense Skip Connections

dc.contributor.authorZhu, Xiaobinen_US
dc.contributor.authorLi, Zhuangzien_US
dc.contributor.authorZhang, Xiaoyuen_US
dc.contributor.authorLi, Haishengen_US
dc.contributor.authorXue, Ziyuen_US
dc.contributor.authorWang, Leien_US
dc.contributor.editorFu, Hongbo and Ghosh, Abhijeet and Kopf, Johannesen_US
dc.date.accessioned2018-10-07T14:59:46Z
dc.date.available2018-10-07T14:59:46Z
dc.date.issued2018
dc.description.abstractRecently, image super-resolution works based on Convolutional Neural Networks (CNNs) and Generative Adversarial Nets (GANs) have shown promising performance. However, these methods tend to generate blurry and over-smoothed super-resolved (SR) images, due to the incomplete loss function and powerless architectures of networks. In this paper, a novel generative adversarial image super-resolution through deep dense skip connections (GSR-DDNet), is proposed to solve the above-mentioned problems. It aims to take advantage of GAN's ability of modeling data distributions, so that GSR-DDNet can select informative feature representation and model the mapping across the low-quality and high-quality images in an adversarial way. The pipeline of the proposed method consists of three main components: 1) The generator of a novel dense skip connection network with the deep structure for learning robust mapping function is proposed to generate SR images from low-resolution images; 2) The feature extraction network based on VGG-19 is adopted to capture high frequency feature maps for content loss; and 3) The discriminator with Wasserstein distance is adopted to identify the overall style of SR and ground-truth images. Experiments conducted on four publicly available datasets demonstrate the superiority against the state-of-the-art methods.en_US
dc.description.number7
dc.description.sectionheadersTowards Better Quality of Images/Videos
dc.description.seriesinformationComputer Graphics Forum
dc.description.volume37
dc.identifier.doi10.1111/cgf.13568
dc.identifier.issn1467-8659
dc.identifier.pages289-300
dc.identifier.urihttps://doi.org/10.1111/cgf.13568
dc.identifier.urihttps://diglib.eg.org:443/handle/10.1111/cgf13568
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.en_US
dc.subjectComputing methodologies
dc.subjectImage processing
dc.subjectComputer systems organization
dc.subjectNeural networks
dc.titleGenerative Adversarial Image Super-Resolution Through Deep Dense Skip Connectionsen_US
Files
Collections