Multi-level Wavelet-CNN for Image Restoration

By Pengju Liu, Hongzhi Zhang, Kai Zhang, Liang Lin, Wangmeng Zuo. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2018

834 citations recorded by the source corpus.

The tradeoff between receptive field size and efficiency is a crucial issue in low level vision. Plain convolutional networks (CNNs) generally enlarge the receptive field at the expense of computational cost. Recently, dilated filtering has been adopted to address this issue. But it suffers from gridding effect, and the resulting receptive field is only a sparse sampling of input image with checkerboard patterns. In this paper, we present a novel multi-level wavelet CNN (MWCNN) model for better tradeoff between receptive field size and computational efficiency. With the modified U-Net architecture, wavelet transform is introduced to reduce the size of feature maps in the contracting subnetwork. Furthermore, another convolutional layer is further used to decrease the channels of feature maps. In the expanding subnetwork, inverse wavelet transform is then deployed to reconstruct the high resolution feature maps. Our MWCNN can also be explained as the generalization of dilated filtering and subsampling, and can be applied to many image restoration tasks. The experimental results clearly show the effectiveness of MWCNN for image denoising, single image super-resolution, and JPEG image artifacts removal.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/5813ae0a48ff50d48f406df52ee5b9795c34a8bd

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Deep Learning for Image Super-Resolution: A Survey (2019, methodology): “[77] feature similarity MS-SSIM [74] multi-scale SSIM IFC [75] information ﬁdelity criterion PSNR peak signal-to-noise ratio IQA image quality assessment SSIM [62] structural similarity MOS mean opinion score VIF [76] visual information ﬁdelity MSSIM [62] mean SSIM BN [140] batch normalization GAN [”
Artificial intelligence in the creative industries: a review (2020, methodology): “[122] developed MWCNN; a similar system that integrates multiscale wavelet transforms within the network to replace max pooling layers in order to better retain visual information.”
Toward Convolutional Blind Denoising of Real Photographs (2018, methodology): “Most recently, other CNN methods, such as RED30 [38], Mem-Net [55], BM3D-Net [60], MWCNN [33] and FFDNet [62], are also developed with promising denoising performance.”
Deep Learning on Image Denoising: An overview (2019, methodology): “(2018) [137] CNN Gaussian image denoisng, super-resolution and JPEG deblocking U-net with wavelet for image restoration Yang et al.”
Efficient and Explicit Modelling of Image Hierarchies for Image Restoration (2023, methodology): “Set QF JPEG DnCNN [92] DCSC [20] QGAC [16] MWCNN [52] FBCNN [32] GRL-S PSNR SSIM PSNR SSIM PSNR SSIM PSNR SSIM PSNR SSIM PSNR SSIM PSNR SSIM”
Attention-guided CNN for image denoising (2020, methodology): “Images C.man House Peppers Starfish Monarch Airplane Parrot Lena Barbara Boat Man Couple Average Noise Level σ = 15 BM3D [8] 31.91 34.93 32.69 31.14 31.85 31.07 31.37 34.26 33.10 32.13 31.92 32.10 32.37 WNNM [13] 32.17 35.13 32.99 31.82 32.71 31.39 31.62 34.27 33.60 32.27 32.11 32.17 32.70 EPLL [64]”
FastDVDnet: Towards Real-Time Deep Video Denoising Without Flow Estimation (2019, background): “Another widespread approach involves the use of convolutional neural networks (CNN), e.g. RBDN [37], MWCNN [30], DnCNN [51], and FFDNet [52].”
SwinIR: Image Restoration Using Swin Transformer (2021, methodology): “Compared methods include traditional models BM3D [14] and WNNM [29], CNN-based models DnCNN [90], IRCNN [91], FFDNet [92], N3Net [65], NLRN [52], FOCNet [38], RNAN [96], MWCNN [54] and DRUNet [88].”
Plug-and-Play Methods Provably Converge with Properly Trained Denoisers (2019, positive): “, 2016), N(3)Net (Plötz & Roth, 2018), and MWCNN (Liu et al., 2018).”
Palette: Image-to-Image Diffusion Models (2021, methodology): “While there are other important prior works on image inpainting, such as [Liu et al. 2018a, 2020; Zheng et al. 2019], we were not able to compare with all of them.”