Deep Wavelet Prediction for Image Super-Resolution

Guo, Tiantong; Mousavi, Hojjat Seyed; Vu, Tiep H.; Monga, Vishal

doi:10.1109/cvprw.2017.148

Cited by 217 publications

(117 citation statements)

References 32 publications

Supporting

Mentioning

115

Contrasting

Unclassified

Order By: Relevance

“…Bae et al [45] proposed a wavelet residual network (WavResNet) with the discovery that CNN learning can benefit from learning on wavelet subbands with features having more channels. For recovering missing details in subbands, Guo et al [46] proposed a deep wavelet super-resolution (DWSR) method. Subsequently, deep convolutional framelets (DCF) [47], [48] had been developed for low-dose CT and inverse problems.…”

Section: A Image Restorationmentioning

confidence: 99%

Multi-Level Wavelet Convolutional Neural Networks

et al. 2019

View full text Add to dashboard Cite

In computer vision, convolutional networks (CNNs) often adopts pooling to enlarge receptive field which has the advantage of low computational complexity. However, pooling can cause information loss and thus is detrimental to further operations such as features extraction and analysis. Recently, dilated filter has been proposed to trade off between receptive field size and efficiency. But the accompanying gridding effect can cause a sparse sampling of input images with checkerboard patterns. To address this problem, in this paper, we propose a novel multi-level wavelet CNN (MWCNN) model to achieve better trade-off between receptive field size and computational efficiency. The core idea is to embed wavelet transform into CNN architecture to reduce the resolution of feature maps while at the same time, increasing receptive field. Specifically, MWCNN for image restoration is based on U-Net architecture, and inverse wavelet transform (IWT) is deployed to reconstruct the high resolution (HR) feature maps. The proposed MWCNN can also be viewed as an improvement of dilated filter and a generalization of average pooling, and can be applied to not only image restoration tasks, but also any CNNs requiring a pooling operation. The experimental results demonstrate effectiveness of the proposed MWCNN for tasks such as image denoising, single image super-resolution, JPEG image artifacts removal and object classification. The code and pre-trained models will be given at https://github.com/lpj-github-io/MWCNNv2. Index Terms-Convolutional networks, receptive field size, efficiency, multi-level wavelet.

show abstract

Section: A Image Restorationmentioning

confidence: 99%

Multi-Level Wavelet Convolutional Neural Networks

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Note that gradient of W l L k,m is not dependent on the first two terms of the loss function in Eq. 14) where Sm a = [s a i−k, j−m ], s a i−k, j−m is the (i − k, j − m) coefficient of Sm. Sh a can also defined in the similar fashion.…”

Section: Dnsp With Data Adaptive Sharpness Enhancing Filtersmentioning

confidence: 99%

“…To speed up the training process, we further extract patches of size 40 × 40 from these bicubic enlarged LR training images. Note that this is also a standard procedure used for training a typical deep SR network [11], [12], [14]. Parameter Choices: To obtain an accurate rank surrogate of Y , we chose δ = .01 based on guidelines mentioned in [40].…”

Section: Experimental Evaluation a Experimental Setupmentioning

confidence: 99%

Deep MR Brain Image Super-Resolution Using Spatio-Structural Priors

Cherukuri

Guo

Schiff

et al. 2020

IEEE Trans. on Image Process.

Self Cite

View full text Add to dashboard Cite

High resolution Magnetic Resonance (MR) images are desired for accurate diagnostics. In practice, image resolution is restricted by factors like hardware and processing constraints. Recently, deep learning methods have been shown to produce compelling state-of-the-art results for image enhancement/superresolution. Paying particular attention to desired hi-resolution MR image structure, we propose a new regularized network that exploits image priors, namely a low-rank structure and a sharpness prior to enhance deep MR image super-resolution (SR). Our contributions are then incorporating these priors in an analytically tractable fashion as well as towards a novel prior guided network architecture that accomplishes the superresolution task. This is particularly challenging for the low rank prior since the rank is not a differentiable function of the image matrix (and hence the network parameters), an issue we address by pursuing differentiable approximations of the rank. Sharpness is emphasized by the variance of the Laplacian which we show can be implemented by a fixed feedback layer at the output of the network. As a key extension, we modify the fixed feedback (Laplacian) layer by learning a new set of training data driven filters that are optimized for enhanced sharpness. Experiments performed on publicly available MR brain image databases and comparisons against existing state-of-the-art methods show that the proposed prior guided network offers significant practical gains in terms of improved SNR/image quality measures. Because our priors are on output images, the proposed method is versatile and can be combined with a wide variety of existing network architectures to further enhance their performance.

show abstract

“…The SRGAN proposed in [29] generates the photorealistic SR images by exploiting the generative adversarial losses. Guo et al [13] proposed a CNN for the SR in the wavelet domain, and have shown that Haar wavelet domain is an efficient one for the SISR. There are also many other structures and methods such as the recursive architecture [23,47] and the Laplacian pyramids [28].…”

Section: Single Image Super-resolution Based On Cnnmentioning

confidence: 99%

Joint High Dynamic Range Imaging and Super-Resolution From a Single Image

2019

View full text Add to dashboard Cite

This paper presents a new framework for jointly enhancing the resolution and the dynamic range of an image, i.e., simultaneous super-resolution (SR) and high dynamic range imaging (HDRI), based on a convolutional neural network (CNN). From the common trends of both tasks, we train a CNN for the joint HDRI and SR by focusing on the reconstruction of high-frequency details. Specifically, the high-frequency component in our work is the reflectance component according to the Retinex-based image decomposition, and only the reflectance component is manipulated by the CNN while another component (illumination) is processed in a conventional way. In training the CNN, we devise an appropriate loss function that contributes to the naturalness quality of resulting images. Experiments show that our algorithm outperforms the cascade implementation of CNN-based SR and HDRI.

show abstract

Deep Wavelet Prediction for Image Super-Resolution

Cited by 217 publications

References 32 publications

Multi-Level Wavelet Convolutional Neural Networks

Multi-Level Wavelet Convolutional Neural Networks

Deep MR Brain Image Super-Resolution Using Spatio-Structural Priors

Joint High Dynamic Range Imaging and Super-Resolution From a Single Image

Contact Info

Product

Resources

About