Open AccessPosted Content
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks
Xintao Wang,Ke Yu,Shixiang Wu,Jinjin Gu,Yihao Liu,Chao Dong,Chen Change Loy,Yu Qiao,Xiaoou Tang +8 more
Reads0
Chats0
TLDR
This work thoroughly study three key components of SRGAN – network architecture, adversarial loss and perceptual loss, and improves each of them to derive an Enhanced SRGAN (ESRGAN), which achieves consistently better visual quality with more realistic and natural textures than SRGAN.Abstract:
The Super-Resolution Generative Adversarial Network (SRGAN) is a seminal work that is capable of generating realistic textures during single image super-resolution. However, the hallucinated details are often accompanied with unpleasant artifacts. To further enhance the visual quality, we thoroughly study three key components of SRGAN - network architecture, adversarial loss and perceptual loss, and improve each of them to derive an Enhanced SRGAN (ESRGAN). In particular, we introduce the Residual-in-Residual Dense Block (RRDB) without batch normalization as the basic network building unit. Moreover, we borrow the idea from relativistic GAN to let the discriminator predict relative realness instead of the absolute value. Finally, we improve the perceptual loss by using the features before activation, which could provide stronger supervision for brightness consistency and texture recovery. Benefiting from these improvements, the proposed ESRGAN achieves consistently better visual quality with more realistic and natural textures than SRGAN and won the first place in the PIRM2018-SR Challenge. The code is available at this https URL .read more
Citations
More filters
Posted Content
Deep learning of multi-resolution X-Ray micro-CT images for multi-scale modelling.
TL;DR: In this article, a 3D Enhanced Deep Super Resolution (EDSR) convolutional neural network was proposed to generate high-resolution data over large spatial scales from low-resolution X-ray data.
Posted Content
Optimizing Generative Adversarial Networks for Image Super Resolution via Latent Space Regularization.
Sheng Zhong,Shifu Zhou +1 more
TL;DR: This paper explicitly applies the Lipschitz Continuity Condition (LCC) to regularize the GAN, and an encoding network that maps the image space to a new optimal latent space is derived from the LCC, and it is used to augment the GAn as a coupling component.
Journal ArticleDOI
Ultra-Fast Laser Fabrication of Alumina Micro-Sample Array and High-Throughput Characterization of Microstructure and Hardness
Xiao Geng,Jianan Tang,Bridget Sheridan,Siddhartha Sarkar,Jianhua Tong,Hai Xiao,Dongsheng Li,Rajendra K. Bordia,Fei Peng +8 more
TL;DR: In this article, the authors demonstrate the ultra-fast fabrication of an alumina sample array and the high-throughput hardness characterization of these sample units using micro-indentation and scanning electron microscopy.
Proceedings ArticleDOI
Comparison of Super-resolution Reconstruction Algorithms Based on Texture Feature Classification
Rui Yang,Wenxi Wang +1 more
TL;DR: By calculating the perceptual index (PI) and root-mean-square error (RMSE) parameters of image sets under different reconstruction models, it can be concluded that enhanced super-resolution generative adversarial network (ESRGAN) has a significant effect on improving the perceived quality of images, while enhanced deep residual networks (EDSR) has the optimal algorithm accuracy.
Posted Content
Deeply Matting-based Dual Generative Adversarial Network for Image and Document Label Supervision.
Yubao Liu,Kai Lin +1 more
TL;DR: A matting-based dual generative adversarial network (mdGAN) for document image SR that uses the input image's corresponding ground truth text label as extra supervise information to refine the two-branch networks during training.
References
More filters
Proceedings ArticleDOI
Deep Residual Learning for Image Recognition
TL;DR: In this article, the authors proposed a residual learning framework to ease the training of networks that are substantially deeper than those used previously, which won the 1st place on the ILSVRC 2015 classification task.
Proceedings Article
Adam: A Method for Stochastic Optimization
Diederik P. Kingma,Jimmy Ba +1 more
TL;DR: This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.
Proceedings Article
Very Deep Convolutional Networks for Large-Scale Image Recognition
Karen Simonyan,Andrew Zisserman +1 more
TL;DR: This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
Proceedings Article
Very Deep Convolutional Networks for Large-Scale Image Recognition
Karen Simonyan,Andrew Zisserman +1 more
TL;DR: In this paper, the authors investigated the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting and showed that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 layers.
Journal ArticleDOI
Generative Adversarial Nets
Ian Goodfellow,Jean Pouget-Abadie,Mehdi Mirza,Bing Xu,David Warde-Farley,Sherjil Ozair,Aaron Courville,Yoshua Bengio +7 more
TL;DR: A new framework for estimating generative models via an adversarial process, in which two models are simultaneously train: a generative model G that captures the data distribution and a discriminative model D that estimates the probability that a sample came from the training data rather than G.