Image denoising using deep CNN with batch renormalization

By Chunwei Tian, Yong Xu, Wangmeng Zuo. Neural Networks, 2020

551 citations recorded by the source corpus.

Deep convolutional neural networks (CNNs) have attracted great attention in the field of image denoising. However, there are two drawbacks: (1) it is very difficult to train a deeper CNN for denoising tasks, and (2) most of deeper CNNs suffer from performance saturation. In this paper, we report the design of a novel network called a batch-renormalization denoising network (BRDNet). Specifically, we combine two networks to increase the width of the network, and thus obtain more features. Because batch renormalization is fused into BRDNet, we can address the internal covariate shift and small mini-batch problems. Residual learning is also adopted in a holistic way to facilitate the network training. Dilated convolutions are exploited to extract more information for denoising tasks. Extensive experimental results show that BRDNet outperforms state-of-the-art image-denoising methods. The code of BRDNet is accessible at http://www.yongxu.org/lunwen.html.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/b876422091c91ef739c958737f3f600a62ec9723

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

1. Yong Xu
2. Chunwei Tian
3. Lunke Fei
4. Hong Liu
5. Zuoyong Li
6. Boyun Li
7. Jiancheng Lv
8. Peng Hu

In Media Technology

1. Chunwei Tian
2. Yong Xu
3. Lunke Fei
4. Hong Liu
5. Zuoyong Li
6. Boyun Li
7. Jiancheng Lv
8. Xiaocui Peng

Citations in the corpus

• Deep Learning on Image Denoising: An overview (2019, unknown): “To address the problem of resource-constraints problem, Tian et al. [207] used a dual CNN with batch renormalization [88], RL and dilated convolutions to deal with real noisy images.”
• All-In-One Image Restoration for Unknown Corruption (2022, methodology): “To this end, we choose five IRSD methods (i.e., BRDNet [40], LPNet [11], FDGAN [7] and MPRNet [51]) and one IRMD method (i.e., DL [8]) as baselines.”
• SwinIR: Image Restoration Using Swin Transformer (2021, unknown): “Dataset σ BM3D [14] DnCNN [90] IRCNN [91] FFDNet [92] DSNet [64] RPCNN [85] BRDNet [71] RNAN [96] RDN [98] IPT [9] DRUNet [88] SwinIR (ours)”
• Efficient Long-Range Attention Network for Image Super-resolution (2022, unknown): “Dataset RPCNN [60] BRDNet [52] DRUNet [63] SwinIR [29] ELAN”
• Attention-guided CNN for image denoising (2020, background): “The MSE denotes the error between a real clean image and a latent clean image [46].”