Lune

CVPR2021Top-tier venue

Pseudo 3D Auto-Correlation Network for Real Image Denoising

Xiaowan Hu, Ruijun Ma, Zhihong Liu, Yuanhao Cai, Xiaole Zhao, Yulun Zhang, Haoqian Wang

Abstract

The extraction of auto-correlation in images has shown great potential in deep learning networks, such as the selfattention mechanism in the channel domain and the selfsimilarity mechanism in the spatial domain. However, the realization of the above mechanisms mostly requires complicated module stacking and a large number of convolution calculations, which inevitably increases model complexity and memory cost. Therefore, we propose a pseudo 3D auto-correlation network (P3AN) to explore a more efficient way of capturing contextual information in image denoising. On the one hand, P3AN uses fast 1D convolution instead of dense connections to realize criss-cross interaction, which requires less computational resources. On the other hand, the operation does not change the feature size and makes it easy to expand. It means that only a simple adaptive fusion is needed to obtain contextual information that includes both the channel domain and the spatial domain. Our method built a pseudo 3D auto-correlation attention block through 1D convolutions and a lightweight 2D structure for more discriminative features. Extensive experiments have been conducted on three synthetic and four real noisy datasets. According to quantitative metrics and visual quality evaluation, the P3AN shows great superiority and surpasses state-of-the-art image denoising methods.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

10 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Xformer: Hybrid X-Shaped Transformer for Image Denoising cites Real Image Denoising With Feature Attention
  • Xformer: Hybrid X-Shaped Transformer for Image Denoising cites Transfer Learning From Synthetic to Real-Noise Denoising With Adaptive Instance Normalization
  • Adaptive Dynamic Filtering Network for Image Denoising cites Real Image Denoising With Feature Attention
  • Adaptive Dynamic Filtering Network for Image Denoising cites Transfer Learning From Synthetic to Real-Noise Denoising With Adaptive Instance Normalization
  • Flow-Guided Sparse Transformer for Video Deblurring cites Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial Training
  • Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial Training cites Real Image Denoising With Feature Attention
  • Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial Training cites Transfer Learning From Synthetic to Real-Noise Denoising With Adaptive Instance Normalization
  • Transfer Learning From Synthetic to Real-Noise Denoising With Adaptive Instance Normalization cites Real Image Denoising With Feature Attention
  • AP-BSN: Self-Supervised Denoising for Real-World Images via Asymmetric PD and Blind-Spot Network cites Transfer Learning From Synthetic to Real-Noise Denoising With Adaptive Instance Normalization
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers10

  1. AP-BSN: Self-Supervised Denoising for Real-World Images via Asymmetric PD and Blind-Spot NetworkWooseok Lee, Sanghyun Son, Kyoung Mu LeeCVPR 2022 · 148 citations
  2. Flow-Guided Sparse Transformer for Video DeblurringJing Lin, Yuanhao Cai, Xiaowan Hu, Haoqian Wang et al.ICML 2022 · 82 citations
  3. Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial TrainingYuanhao Cai, Xiaowan Hu, Haoqian Wang, Yulun Zhang et al.NeurIPS 2021 · 81 citations
  4. Adaptive Dynamic Filtering Network for Image DenoisingHao Shen, Zhong-Qiu Zhao, Wandi ZhangAAAI 2023 · 69 citations
  5. Xformer: Hybrid X-Shaped Transformer for Image DenoisingJiale Zhang, Yulun Zhang, Jinjin Gu, Jiahua Dong et al.ICLR 2024 · 59 citations

Builds on3

  1. CCNet: Criss-Cross Attention for Semantic SegmentationZilong Huang, Xinggang Wang, Lichao Huang, Chang Huang et al.ICCV 2019 · 2,972 citations
  2. Real Image Denoising With Feature AttentionSaeed Anwar, Nick BarnesICCV 2019 · 644 citations
  3. Transfer Learning From Synthetic to Real-Noise Denoising With Adaptive Instance NormalizationYoonsik Kim, Jae Woong Soh, Gu Yong Park, Nam Ik ChoCVPR 2020

Related papers

  1. Dual-Domain Attention for Image DeblurringYuning Cui, Yi Tao, Wenqi Ren, Alois KnollAAAI 2023 · 68 citations
  2. Generative Adaptive Convolutions for Real-World Noisy Image DenoisingRuijun Ma, Shuyi Li, Bob Zhang, Zhengming LiAAAI 2022 · 34 citations
  3. Attention Cube Network for Image RestorationYucheng Hang, Qingmin Liao, Wenming Yang, Yupeng Chen et al.ACM MM 2020 · 22 citations
  4. ECA-Net: Efficient Channel Attention for Deep Convolutional Neural NetworksQilong Wang, Banggu Wu, Pengfei Zhu, Peihua Li et al.CVPR 2020
  5. EDNet: Efficient Disparity Estimation With Cost Volume Combination and Attention-Based Spatial ResidualSongyan Zhang, Zhicheng Wang, Qiang Wang, Jinshuo Zhang et al.CVPR 2021
Dusk over the sea between two cliffs drawn in fine vertical lines