Image Super-Resolution via Dual-State Recurrent Networks

Han, Wei; Chang, Shiyu; Liu, Ding; Yu, Mei; Witbrock, Michael; Huang, Thomas S.

doi:10.1109/cvpr.2018.00178

Cited by 211 publications

(142 citation statements)

References 42 publications

Supporting

Mentioning

142

Contrasting

Order By: Relevance

“…PSNR, SSIM). While the first attempt to apply a conv-net to image SR [6] did not significantly surpass the performance of prior methods, it set the ground for major improvements in PSNR/SSIM values over the course of the several following years [15,17,18,19,39,11,52,34,10,51]. During these years, the rising PSNR/SSIM values were not always accompanied by a rise in the perceptual quality.…”

Section: Perceptual Super Resolutionmentioning

confidence: 97%

The 2018 PIRM Challenge on Perceptual Image Super-Resolution

Blau

Mechrez

Timofte

et al. 2019

Lecture Notes in Computer Science

414

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

Section: Perceptual Super Resolutionmentioning

confidence: 97%

The 2018 PIRM Challenge on Perceptual Image Super-Resolution

Blau

Mechrez

Timofte

et al. 2019

Lecture Notes in Computer Science

414

318

View full text Add to dashboard Cite

show abstract

“…Wang et al [28] proposed a novel deep spatial feature transform to recover textures conditioned on the categorical priors. Both DBPN [10] and DSRN [9] made use of the mutual dependencies of low-and high-resolution images. DBPN exploited iterative up-sampling and down-sampling layers to provide an error feedback mechanism for each stage.…”

Section: Single Image Super Resolutionmentioning

confidence: 99%

Meta-SR: A Magnification-Arbitrary Network for Super-Resolution

Zhang³

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

389

312

View full text Add to dashboard Cite

Recent research on super-resolution has achieved great success due to the development of deep convolutional neural networks (DCNNs). However, super-resolution of arbitrary scale factor has been ignored for a long time. Most previous researchers regard super-resolution of different scale factors as independent tasks. They train a specific model for each scale factor which is inefficient in computing, and prior work only take the super-resolution of several integer scale factors into consideration. In this work, we propose a novel method called Meta-SR to firstly solve super-resolution of arbitrary scale factor (including noninteger scale factors) with a single model. In our Meta-SR, the Meta-Upscale Module is proposed to replace the traditional upscale module. For arbitrary scale factor, the Meta-Upscale Module dynamically predicts the weights of the upscale filters by taking the scale factor as input and use these weights to generate the HR image of arbitrary size. For any low-resolution image, our Meta-SR can continuously zoom in it with arbitrary scale factor by only using a single model. We evaluated the proposed method through extensive experiments on widely used benchmark datasets on single image super-resolution. The experimental results show the superiority of our Meta-Upscale.

show abstract

“…Han et al [38] considered the DRCN [17] and the Deep Recursive Residual Network (DRRN) [21] as the Recurrent Neural Networks (RNNs) employing recurrent states and proposed Dual-State Recurrent Network (DSRN) [38], which uses dual recurrent states.…”

Section: Introductionmentioning

confidence: 99%

Multi-Scale Inception Based Super-Resolution Using Deep Learning Approach

Muhammad¹,

Aramvith²

2019

Electronics

View full text Add to dashboard Cite

Single image super-resolution (SISR) aims to reconstruct a high-resolution (HR) image from a low-resolution (LR) image. In order to address the SISR problem, recently, deep convolutional neural networks (CNNs) have achieved remarkable progress in terms of accuracy and efficiency. In this paper, an innovative technique, namely a multi-scale inception-based super-resolution (SR) using deep learning approach, or MSISRD, was proposed for fast and accurate reconstruction of SISR. The proposed network employs the deconvolution layer to upsample the LR image to the desired HR image. The proposed method is in contrast to existing approaches that use the interpolation techniques to upscale the LR image. Primarily, interpolation techniques are not designed for this purpose, which results in the creation of undesired noise in the model. Moreover, the existing methods mainly focus on the shallow network or stacking multiple layers in the model with the aim of creating a deeper network architecture. The technique based on the aforementioned design creates the vanishing gradients problem during the training and increases the computational cost of the model. Our proposed method does not use any hand-designed pre-processing steps, such as the bicubic interpolation technique. Furthermore, an asymmetric convolution block is employed to reduce the number of parameters, in addition to the inception block adopted from GoogLeNet, to reconstruct the multiscale information. Experimental results demonstrate that the proposed model exhibits an enhanced performance compared to twelve state-of-the-art methods in terms of the average peak signal-to-noise ratio (PSNR), structural similarity index (SSIM) with a reduced number of parameters for the scale factor of 2×, 4×, and 8×. blurry results in smooth region areas. Furthermore, reconstruction-based methods are very efficient in preserving sharp edges or boundaries and suppressing the jagged ringing artifacts [10]. To reconstruct HR images with complex scenes, prior methods fail to reconstruct the high-frequency information details. Recently, approaches have been used to learn the nonlinear mapping of image space between LR to HR images through millions of image pair co-occurrences, comprising linear regression [11], spare based dictionary learning [12], random forest [13], and neural networks [14,15].Among them, the neural networks have marked-out significant consideration due to easy, simple elements and excellent performance, but there are still some limitations. First, this type of model is a fully deep neural network and is used to limit the contextual information over all the global image region. Although some methods [16][17][18][19][20][21] have revised and improved the image restoration quality by stacking side by side convolution layers to display contextual information over a large region but increase the computational cost, as well as memory usage. Second, existing approaches only optimize the network model in the loss function, due to the increase in the blurry edges in the recov...

show abstract

Image Super-Resolution via Dual-State Recurrent Networks

Cited by 211 publications

References 42 publications

The 2018 PIRM Challenge on Perceptual Image Super-Resolution

The 2018 PIRM Challenge on Perceptual Image Super-Resolution

Meta-SR: A Magnification-Arbitrary Network for Super-Resolution

Multi-Scale Inception Based Super-Resolution Using Deep Learning Approach

Contact Info

Product

Resources

About