Multi-Scale Recursive and Perception-Distortion Controllable Image Super-Resolution
Pablo Navarrete Michelini,Dan Zhu,Hanwen Liu +2 more
- pp 3-19
TLDR
In this paper, a discriminator for adversarial training is proposed, which is multi-scale that resembles a progressive-GAN and includes a new layer to capture significant statistics of natural images.Abstract:
We describe our solution for the PIRM Super–Resolution Challenge 2018 where we achieved the \(\varvec{2^{nd}}\) best perceptual quality for average \(RMSE\leqslant 16\), \(5^{th}\) best for \(RMSE\leqslant 12.5\), and \(7^{th}\) best for \(RMSE\leqslant 11.5\). We modify a recently proposed Multi–Grid Back–Projection (MGBP) architecture to work as a generative system with an input parameter that can control the amount of artificial details in the output. We propose a discriminator for adversarial training with the following novel properties: it is multi–scale that resembles a progressive–GAN; it is recursive that balances the architecture of the generator; and it includes a new layer to capture significant statistics of natural images. Finally, we propose a training strategy that avoids conflicts between reconstruction and perceptual losses. Our configuration uses only 281 k parameters and upscales each image of the competition in 0.2 s in average.read more
Citations
More filters
Book ChapterDOI
The 2018 PIRM Challenge on Perceptual Image Super-Resolution
TL;DR: This paper reports on the 2018 PIRM challenge on perceptual super-resolution (SR), held in conjunction with the Perceptual Image Restoration and Manipulation (PIRM) workshop at ECCV 2018, and concludes with an analysis of the current trends in perceptual SR, as reflected from the leading submissions.
Proceedings Article
PIPAL: a Large-Scale Image Quality Assessment Dataset for Perceptual Image Restoration
TL;DR: It is indicated that existing IQA methods cannot fairly evaluate GAN-based IR algorithms, and a large-scale IQA dataset, called Perceptual Image Processing Algorithms (PIPAL) is contributed, which includes the results of GAn-based methods, which are missing in previous datasets.
Proceedings ArticleDOI
Explorable Super Resolution
Yuval Bahat,Tomer Michaeli +1 more
TL;DR: A novel module is proposed that can wrap any existing SR network, analytically guaranteeing that its SR outputs would precisely match the LR input, when down- sampled, and is guaranteed to decrease the reconstruction error of any SR network it wraps.
Proceedings ArticleDOI
Wavelet Domain Style Transfer for an Effective Perception-Distortion Tradeoff in Single Image Super-Resolution
TL;DR: A novel method based on wavelet domain style transfer (WDST), which achieves a better PD tradeoff than the GAN based methods and achieves the best trade-off between the distortion and perceptual quality among the existing state-of-the-art SISR methods.
Proceedings ArticleDOI
CFSNet: Toward a Controllable Feature Space for Image Restoration
TL;DR: The proposed framework, named Controllable Feature Space Network (CFSNet), is entangled by two branches based on different objectives, which can adaptively learn the coupling coefficients of different layers and channels, which provides finer control of the restored image quality.
References
More filters
Journal ArticleDOI
Image quality assessment: from error visibility to structural similarity
TL;DR: In this article, a structural similarity index is proposed for image quality assessment based on the degradation of structural information, which can be applied to both subjective ratings and objective methods on a database of images compressed with JPEG and JPEG2000.
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.
Proceedings ArticleDOI
Densely Connected Convolutional Networks
TL;DR: DenseNet as mentioned in this paper proposes to connect each layer to every other layer in a feed-forward fashion, which can alleviate the vanishing gradient problem, strengthen feature propagation, encourage feature reuse, and substantially reduce the number of parameters.
Proceedings ArticleDOI
Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks
TL;DR: CycleGAN as discussed by the authors learns a mapping G : X → Y such that the distribution of images from G(X) is indistinguishable from the distribution Y using an adversarial loss.
Posted Content
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
TL;DR: This work introduces a class of CNNs called deep convolutional generative adversarial networks (DCGANs), that have certain architectural constraints, and demonstrates that they are a strong candidate for unsupervised learning.