Perception-Enhanced Image Super-Resolution via Relativistic Generative Adversarial Networks

Vu, Thang; Luu, Tung M.; Yoo, Chang D.

doi:10.1007/978-3-030-11021-5_7

Cited by 30 publications

(37 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Benchmarks. The comparison methods are classified into three categories: methods that aim to improve the objective quality including A+ [25], Self-Ex [9], SRCNN [6], ESPCN [22], SRResNet-MSE [14], VDSR [12], EDSR [15], and RCAN [30]; methods that aim to improve the perceptual quality including SRGAN-vgg54 [14], SRGAN-vgg22 [14], ENet [21], and CX [19]; and methods that aim to improve both the objective and perceptual quality including SRGAN-MSE [14], GMGBP [20], PESR [26], EUSR [4], Deng [5] and ESRGAN [27].…”

Section: Numerical Resultsmentioning

confidence: 99%

“…Among all these methods, our method is the closest to the 24 bottom left corner, which means that we achieve the best trade-off between the objective and perceptual quality. Table 2 compares the numerical results of our method with SRGAN-MSE [14], GMGBP [20] , PESR [26], Deng [5] and ESRGAN [27] (with α =0.8), which all aim to improve both the perceptual and objective quality. As we can see, our method outperforms others in both perceptual and objective quality.…”

Section: Numerical Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Wavelet Domain Style Transfer for an Effective Perception-Distortion Tradeoff in Single Image Super-Resolution

Deng

Yang

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

In single image super-resolution (SISR), given a lowresolution (LR) image, one wishes to find a high-resolution (HR) version of it which is both accurate and photorealistic. Recently, it has been shown that there exists a fundamental tradeoff between low distortion and high perceptual quality [3], and the generative adversarial network (GAN) is demonstrated to approach the perceptiondistortion (PD) bound effectively. In this paper, we propose a novel method based on wavelet domain style transfer (WDST), which achieves a better PD tradeoff than the GAN based methods. Specifically, we propose to use 2D stationary wavelet transform (SWT) to decompose one image into low-frequency and high-frequency sub-bands. For the low-frequency sub-band, we improve its objective quality through an enhancement network. For the high-frequency sub-band, we propose to use WDST to effectively improve its perceptual quality. By feat of the perfect reconstruction property of wavelets, these sub-bands can be re-combined to obtain an image which has simultaneously high objective and perceptual quality. The numerical results on various datasets show that our method achieves the best trade-off between the distortion and perceptual quality among the existing state-of-the-art SISR methods.

show abstract

Section: Numerical Resultsmentioning

confidence: 99%

Section: Numerical Resultsmentioning

confidence: 99%

Wavelet Domain Style Transfer for an Effective Perception-Distortion Tradeoff in Single Image Super-Resolution

Deng

Yang

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

show abstract

“…Image super-resolution reconstruction based on deep learning has developed rapidly. Most of the algorithms based on deep learning mainly improve network structure [8,10,12,19] and loss function [9,11,25,26,28,34]. In addition, to explore the dependencies between feature maps, many algorithms introduce attention mechanisms [22][23][24].…”

Section: Related Workmentioning

confidence: 99%

“…In addition, the proposed discriminative losses [9,18] can also improve the perceptual quality of SR images. A relative discriminator [27] has also been used in image super-resolution [14,28]. Vu et al [28] enhanced RaGAN by wrapping the focal loss.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Generative Adversarial Network-Based Super-Resolution Considering Quantitative and Perceptual Quality

Wang

Cheng

et al. 2020

Symmetry

View full text Add to dashboard Cite

In recent years, the common algorithms for image super-resolution based on deep learning have been increasingly successful, but there is still a large gap between the results generated by each algorithm and the ground-truth. Even some algorithms that are dedicated to image perception produce more textures that do not exist in the original image, and these artefacts also affect the visual perceptual quality of the image. We believe that in the existing perceptual-based image super-resolution algorithm, it is necessary to consider Super-Resolution (SR) image quality, which can restore the important structural parts of the original picture. This paper mainly improves the Enhanced Super-Resolution Generative Adversarial Networks (ESRGAN) algorithm in the following aspects: adding a shallow network structure, adding the dual attention mechanism in the generator and the discriminator, including the second-order channel mechanism and spatial attention mechanism and optimizing perceptual loss by adding second-order covariance normalization at the end of feature extractor. The results of this paper ensure image perceptual quality while reducing image distortion and artefacts, improving the perceived similarity of images and making the images more in line with human visual perception.

show abstract

The 2018 PIRM Challenge on Perceptual Image Super-Resolution

Blau

Mechrez

Timofte

et al. 2019

Lecture Notes in Computer Science

410

318

View full text Add to dashboard Cite

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. In contrast to previous SR challenges, our evaluation methodology jointly quantifies accuracy and perceptual quality, therefore enabling perceptualdriven methods to compete alongside algorithms that target PSNR maximization. Twenty-one participating teams introduced algorithms which well-improved upon the existing state-of-the-art methods in perceptual SR, as confirmed by a human opinion study. We also analyze popular image quality measures and draw conclusions regarding which of them correlates best with human opinion scores. We conclude with an analysis of the current trends in perceptual SR, as reflected from the leading submissions.

show abstract

Perception-Enhanced Image Super-Resolution via Relativistic Generative Adversarial Networks

Cited by 30 publications

References 45 publications

Wavelet Domain Style Transfer for an Effective Perception-Distortion Tradeoff in Single Image Super-Resolution

Wavelet Domain Style Transfer for an Effective Perception-Distortion Tradeoff in Single Image Super-Resolution

Generative Adversarial Network-Based Super-Resolution Considering Quantitative and Perceptual Quality

The 2018 PIRM Challenge on Perceptual Image Super-Resolution

Contact Info

Product

Resources

About