Multi-Scale Progressive Fusion Network for Single Image Deraining

  title={Multi-Scale Progressive Fusion Network for Single Image Deraining},
  author={Kui Jiang and Zhongyuan Wang and Peng Yi and Chen Chen and Baojin Huang and Yimin Luo and Jiayi Ma and Junjun Jiang},
  journal={2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
Rain streaks in the air appear in various blurring degrees and resolutions due to different distances from their positions to the camera. Similar rain patterns are visible in a rain image as well as its multi-scale (or multi-resolution) versions, which makes it possible to exploit such complementary information for rain streak representation. In this work, we explore the multi-scale collaborative representation for rain streaks from the perspective of input image scales and hierarchical deep… 

Multi-Scale Hourglass Hierarchical Fusion Network for Single Image Deraining

Single image deraining has become a vital preprocessing step in the practical applications for object detection, recognition, and tracking tasks under rainy situations.

Single image deraining using multi‐stage and multi‐scale joint channel coordinate attention fusion network

This work proposes an effective algorithm, called multi‐stage and multi‐scale joint channel coordinate attention fusion network (MMAFN), both of which use an encoder‐decoder network to extract features and integrates the features of the former to further refine features.

Multi-Scale Channel Transformer Network for Single Image Deraining

This study proposes a novel Channel Transformer, which performs self-attention in the channel direction instead of the spatial direction, and first incorporates multiple channel transformer blocks into a multi- scale architecture to extract multi-scale contexts and exploit channel long-dependence.

Multi-scale Residual Aggregation Deraining Network with Spatial Context-aware Pooling and Activation

This study proposes a novel multi-scale residual aggregation network, to effectively solve the single image deraining problem and proposes the spatial context-aware pooling and activation method (SCAP and activation) for incorporating with the authors' deraining network to boost performance.

Contrastive learning-based generative network for single image deraining

An innovative contrastive learning strategy is proposed and applied to each stage of the proposed CLGNet to enhance the decoupling ability of the encoder and help the model recognize complex rain patterns.

Multifocal Attention-Based Cross-Scale Network for Image De-raining

An effective multifocal attention-based cross-scale network is constructed, which exhaustively utilizes the cross- scale correlations of both rain streaks and background, to achieve image de-raining.

Multistage Feature Complimentary Network for Single-Image Deraining

A multistage framework based on progressive restoration combined with recurrent neural network and feature complementarity technology to remove rain streak from single images can remove rain more completely, preserve more background details, and achieve better visual effects compared with some popular single-image deraining methods.

Cascaded Attention Guidance Network for Single Rainy Image Restoration

An advanced cascaded attention guidance network, dubbed as CAG-Net, is formulated and designed as a three-stage model, and is demonstrated to produce significantly better results than state-of-the-art models.

MARA-Net: Single Image Deraining Network with Multi-level connection and Adaptive Regional Attention

A multi-level connection and adaptive regional attention network (MARA-Net) to properly restore the original background textures in rainy images and demonstrates that the proposed model significantly outperforms existing state-of-the-art models.

Decomposition Makes Better Rain Removal: An Improved Attention-Guided Deraining Network

An improved non-local block is designed to exploit the self-similarity of rain information by learning the holistic spatial feature correlations while reducing the calculation complexity.



Single Image Deraining using a Recurrent Multi-scale Aggregation and Enhancement Network

  • Youzhao YangHong Lu
  • Computer Science
    2019 IEEE International Conference on Multimedia and Expo (ICME)
  • 2019
A novel progressive single image deraining method named Recurrent Multi-scale Aggregation and Enhancement Network (ReMAEN), which contains a symmetric structure where recurrent blocks with shared channel attention are applied to select useful information collaboratively and remove rain streaks stage by stage.

Single Image Deraining using Scale-Aware Multi-Stage Recurrent Network

A scale-aware multi-stage convolutional neural network that treats the same class of objects according to their unique sub-classes is novel, particularly in the context of rain removal, and is effective and outperforms the state-of-the-art methods.

Recurrent Squeeze-and-Excitation Context Aggregation Net for Single Image Deraining

A novel deep network architecture based on deep convolutional and recurrent neural networks for single image deraining based on contextual information is proposed and outperforms the state-of-the-art approaches under all evaluation metrics.

Scale-Free Single Image Deraining Via Visibility-Enhanced Recurrent Wavelet Learning

To handle heavy rain cases where rain streak accumulation is presented, a detail appearing rain accumulation removal is constructed to not only improve the visibility but also enhance the details in dark regions.

Uncertainty Guided Multi-Scale Residual Learning-Using a Cycle Spinning CNN for Single Image De-Raining

The proposed Uncertainty guided Multi-scale Residual Learning (UMRL) network attempts to address the single image de-raining problem by learning the rain content at different scales and using them to estimate the final de-rained output.

Residual Multiscale Based Single Image Deraining

This paper proposes a residual multiscale pyramid based single image deraining method to alleviate the difficulty of rain image decomposition and demonstrates that this method outperforms the state of the art significantly.

Heavy Rain Image Restoration: Integrating Physics Model and Conditional Adversarial Learning

A physics-based backbone followed by a depth-guided GAN refinement to recover the background details failed to be retrieved by the first stage, as well as correcting artefacts introduced by that stage.

Density-Aware Single Image De-raining Using a Multi-stream Dense Network

A novel density-aware multi-stream densely connected convolutional neural network-based algorithm, called DID-MDN, for joint rain density estimation and de-raining, which achieves significant improvements over the recent state-of-the-art methods.

Deep Joint Rain Detection and Removal from a Single Image

A recurrent rain detection and removal network that removes rain streaks and clears up the rain accumulation iteratively and progressively is proposed and a new contextualized dilated network is developed to exploit regional contextual information and to produce better representations for rain detection.

Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution

This paper proposes the Laplacian Pyramid Super-Resolution Network (LapSRN) to progressively reconstruct the sub-band residuals of high-resolution images and generates multi-scale predictions in one feed-forward pass through the progressive reconstruction, thereby facilitates resource-aware applications.