Unpaired Deep Image Deraining Using Dual Contrastive Learning

Chen, Xiang; Pan, Jinshan; Jiang, Kui; Li, Yufeng; Huang, Yufeng; Kong, Caihua; Dai, Longgang; Fan, Zhentao

doi:10.1109/cvpr52688.2022.00206

Cited by 129 publications

(31 citation statements)

References 131 publications

(317 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The network can generate clean, high-quality, rain-free images directly regardless of any rainfall assumptions. In 2022 Chen et al [38] proposed an adversarial network DCD-GAN that does not require training on paired rain/no rain datasets to better explore image features by employing double contrast learning. In the same year, Woo et al [39] proposed an end-to-end deraining network and introduced novel dilation-wise and skip attention modules.…”

Section: Deep Network-based Deraining Methodsmentioning

confidence: 99%

See 1 more Smart Citation

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

Section: Deep Network-based Deraining Methodsmentioning

confidence: 99%

“…In 2022 Chen et al. [38] proposed an adversarial network DCD‐GAN that does not require training on paired rain/no rain datasets to better explore image features by employing double contrast learning. In the same year, Woo et al.…”

Section: Related Workmentioning

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

show abstract

“…Chen et al. [29] introduce contrastive constraints in an improved CycleGAN architecture for image deraining. The contrastive features are extracted from encoding layers of generators.…”

Section: Related Workmentioning

confidence: 99%

A domain translation network with contrastive constraint for unpaired motion image deblurring

Zhao

2023

IET Image Processing

View full text Add to dashboard Cite

Most motion deblurring methods require a large amount of paired training data, which is nearly unreachable in practice. To overcome the limitation, a domain translation network with contrastive constraint for unpaired motion image deblurring is proposed. First, a domain translation network with two streams, a sharp domain translation stream and a blurred domain translation stream, to handle unpaired sharp and blurred images from the real world is presented. Second, a contrastive constraint loss in the deep intermediate level for the two streams to promote the network to produce deblurring results close to the real sharp image is proposed. Third, distinct loss functions for the two streams to preserve the edge and texture detail of the deblurring image is designed. Extensive experiments on several benchmark datasets demonstrate that the proposed network achieves better visual performance than the current state‐of‐the‐art methods for unpaired motion image deblurring.

show abstract

“…Similar to supervised metric learning, the self-supervised model learns similar representations from the positive pairs and should be different than the representations of the negative pairs. Various types of metric/contrastive learning works have been developed for image pattern recognition applications, including image classification [6], [11]- [13], image clustering [14], [15], image segmentation [16], [17], image reconstruction [18], [19], and object detection [20]. All these works used geometric proximity (e.g., Cosine similarity or Euclidean distance) as the proximity function in the training an objective loss to learn the geometric representation of the samples.…”

Section: Introductionmentioning

confidence: 99%

Learning Generalized Hybrid Proximity Representation for Image Recognition

Li¹,

Ralescu²

2023

Preprint

View full text Add to dashboard Cite

Recently, deep metric learning techniques received attentions, as the learned distance representations are useful to capture the similarity relationship among samples and further improve the performance of various of supervised or unsupervised learning tasks. We propose a novel supervised metric learning method that can learn the distance metrics in both geometric and probabilistic space for image recognition. In contrast to the previous metric learning methods which usually focus on learning the distance metrics in Euclidean space, our proposed method is able to learn better distance representation in a hybrid approach. To achieve this, we proposed a Generalized Hybrid Metric Loss (GHM-Loss) to learn the general hybrid proximity features from the image data by controlling the tradeoff between geometric proximity and probabilistic proximity. To evaluate the effectiveness of our method, we first provide theoretical derivations and proofs of the proposed loss function, then we perform extensive experiments on two public datasets to show the advantage of our method compared to other stateof-the-art metric learning methods.

show abstract

Unpaired Deep Image Deraining Using Dual Contrastive Learning

Cited by 129 publications

References 131 publications

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

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

A domain translation network with contrastive constraint for unpaired motion image deblurring

Learning Generalized Hybrid Proximity Representation for Image Recognition

Contact Info

Product

Resources

About