Abstract:Rain streaks, which can be viewed as a type of noise, have the potential to destroy the structural integrity of objects within an image, leading to a dramatic degradation in the performance of advanced vision algorithms. Therefore, it is important to develop an effective deraining algorithm. There are three challenges in the current deraining task: First, it is difficult to separate rain streaks from the image background because of the complexity of rain types. Second, most of the existing methods use encoder–decoder-based structures, but the models are designed to be too bulky and inflexible to accommodate complex rain streaks. Third, many models perform the same operation for all rain types, which can reduce the generalizability of the model. To address these challenges, we propose a scale constraint iterative update network which divides the network structure into feature extraction block, multi-scale constraint block, and iterative update block. In the feature extraction block, the deep mixed attention is used to extract complex rain streak features, and the mixed convolution block is used to extract the background details. The multi-scale constraint block aggregates contextual features across different scales, while the Global GateUnit captures long-range dependencies. We followed a coarse-to-fine pipeline. In the iterative update block, the feature maps are updated based on features from different scales, gradually refining the image. The final optimized result is obtained by performing learnable up-sampling. During the whole iteration process, the parameters are shared to reduce the complexity of the model while maintaining the performance. Additionally, We also designed a dynamic inference mechanism, which dynamically selects the number of iterations for different types of rain streaks, effectively enhancing the generalization of the model. We conducted extensive experiments on synthetic and real-world rain datasets, as well as a raindrop dataset. The results of our experiments demonstrate that our method outperforms existing deraining methods in terms of effectiveness and generalization ability.

APAN: Across-scale Progressive Attention Network for Single Image Deraining

Combining multiscale learning and attention mechanism densely connected network for single image deraining

Combined with Pyramid Split Attention and Multi-Scale Feature Learning Network for Single Image Deraining

Enhanced Attentive Generative Adversarial Network for Single-Image Deraining.

Single Image Deraining Via Deep Pyramid Network with Spatial Contextual Information Aggregation

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

Multi-Scale Progressive Fusion Network for Single Image Deraining

Single image deraining using scale constraint iterative update network

Single Image Deraining Via Detail-Guided Efficient Channel Attention Network

Single Image Deraining Via Scale-space Invariant Attention Neural Network.

Single-Image Deraining via Recurrent Residual Multiscale Networks

PFDN: Pyramid Feature Decoupling Network for Single Image Deraining

Online-updated High-order Collaborative Networks for Single Image Deraining

Rain-Free and Residue Hand-in-Hand: A Progressive Coupled Network for Real-Time Image Deraining

Progressive Hybrid-Modulated Network for Single Image Deraining

Attentive Feature Refinement Network for Single Rainy Image Restoration

Deep Image Deraining Via Intrinsic Rainy Image Priors and Multi-scale Auxiliary Decoding

Feature decoupling and reorganization network for single image deraining

Attention-Guided Deraining Network Via Stage-Wise Learning.

Dilated Residual Spatial Attentive Generative Adversarial Network for Single Image De-raining

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