Cumulative Rain Density Sensing Network for Single Image Derain

Peng, Long; Jiang, Aiwen; Yi, Qiaosi; Wang, Mingwen

doi:10.1109/lsp.2020.2974691

Cited by 22 publications

(7 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Structural similarity (SSIM) is a normalized measurement based on perception model, which is used to determine how similar two photos are [15,24].…”

Section: 21ssimmentioning

confidence: 99%

“…Peng et al suggested an adaptive rainstreak removal CRDNet. For cumulative rain density classification, the novel model combines an efficient W-Net with great learning capacity to recover rain invariant low-frequency signals and a cost-sensitive label-encoding technique with improved rain streak separation [24].…”

Section: Yang Et Al Developed a Model In Which One Component Reflects...mentioning

confidence: 99%

“…Did-MDN dataset contains 12000 training images with rain-density labels and 1200 test images [16,24]. NYU-Rain data set takes into account the actual situation that Rain and fog will lead to image blur, and uses images in NYU-DEPthV2 data set as background to provide depth information to generate Rain fringe layer and Rain aggregation layer [18,25].…”

mentioning

confidence: 99%

See 2 more Smart Citations

Deep learning based single image deraining: datasets, metrics and methods

Liu¹

2023

International Conference on Cloud Computing, Performance Computing, and Deep Learning (CCPCDL 2023)

View full text Add to dashboard Cite

“…Structural similarity (SSIM) is a normalized measurement based on perception model, which is used to determine how similar two photos are [15,24].…”

Section: 21ssimmentioning

confidence: 99%

Section: Yang Et Al Developed a Model In Which One Component Reflects...mentioning

confidence: 99%

See 1 more Smart Citation

Deep learning based single image deraining: datasets, metrics and methods

Liu¹

2023

International Conference on Cloud Computing, Performance Computing, and Deep Learning (CCPCDL 2023)

View full text Add to dashboard Cite

“…At the same time, different deraining networks are suggested being composed of various types of modules, including null convolution [15], dense convolution [16,17], residual modules [18], and recurrent layers [19]. The method's effectiveness is further improved by adding various channels and spatial attention mechanisms [20,21].…”

Section: O(x) = B(x) + R(x)mentioning

confidence: 99%

Continuous learning deraining network based on residual FFT convolution and contextual transformer module

Zhang

Peng

et al. 2022

IET Image Processing

View full text Add to dashboard Cite

Single image deraining work has attracted a lot of attention. So an effective single image deraining network model, RFCTNet, is designed. Specifically, the residual Fast Fourier Transform convolution module is used as a baseline for the deraining network. The advantage is its ability to capture both long and short‐term information interactions and to integrate high and low‐frequency background information for delivery. And a novel Transformer module, the Contextual Transformer module, is introduced. Contextual knowledge of the rain streaks neighbourhood is learned to improve the recovery of textures and information in the image background based on focusing on the rain pattern features. Because of the overlap between rain streaks, removing them in one stage is difficult, so a multi‐stage progressive rain removal method is used. Considering the catastrophic forgetting problem of deep learning, a continuous learning method, PIGWM, is adopted, which is c and enables the network model to memorize of the previous data set. Also, the rain removal method RFCTNet has been proven through extensive experiments to outperform other advanced methods' evaluation metrics on both synthetic and real datasets. The source code can be found on https://github.com/SUTwu/RFCTNet

show abstract

“…Recently, convolutional neural networks (CNN) have achieved significant success in many computer vision tasks [32,13,18,29]. Moreover, many CNN-based methods [7,8,43,23,49,15,38,50,31,45,6,37,16] have been proposed for rain removal, such as DDN [8], RES-CAN [23], PReNet [31], DRDNet [6], and RCDNet [37].…”

Section: Introductionmentioning

confidence: 99%

Structure-Preserving Deraining with Residue Channel Prior Guidance

Yi¹,

Li²,

Dai³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Single image deraining is important for many high-level computer vision tasks since the rain streaks can severely degrade the visibility of images, thereby affecting the recognition and analysis of the image. Recently, many CNN-based methods have been proposed for rain removal. Although these methods can remove part of the rain streaks, it is difficult for them to adapt to real-world scenarios and restore high-quality rain-free images with clear and accurate structures. To solve this problem, we propose a Structure-Preserving Deraining Network (SPDNet) with RCP guidance. SPDNet directly generates high-quality rain-free images with clear and accurate structures under the guidance of RCP but does not rely on any rain-generating assumptions. Specifically, we found that the RCP of images contains more accurate structural information than rainy images. Therefore, we introduced it to our deraining network to protect structure information of the rain-free image. Meanwhile, a Wavelet-based Multi-Level Module (WMLM) is proposed as the backbone for learning the background information of rainy images and an Interactive Fusion Module (IFM) is designed to make full use of RCP information. In addition, an iterative guidance strategy is proposed to gradually improve the accuracy of RCP, refining the result in a progressive path. Extensive experimental results on both synthetic and real-world datasets demonstrate that the proposed model achieves new state-of-the-art results. Code: https://github.com/Joyies/SPDNet

show abstract

Cumulative Rain Density Sensing Network for Single Image Derain

Cited by 22 publications

References 20 publications

Deep learning based single image deraining: datasets, metrics and methods

Deep learning based single image deraining: datasets, metrics and methods

Continuous learning deraining network based on residual FFT convolution and contextual transformer module

Structure-Preserving Deraining with Residue Channel Prior Guidance

Contact Info

Product

Resources

About