Progressive Image Deraining Networks: A Better and Simpler Baseline

2019/01/26 by Dongwei Ren, Wangmeng Zuo, Ren, Dongwei +7 · 34 citations
Computer Science · #Advanced Image Processing Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Image Enhancement Techniques #Image and Signal Denoising Methods #cs.CV

paper · pdf · doi:10.48550/arxiv.1901.09221

Accepted to CVPR 2019. The codes, pre-trained models and results are available at https://github.com/csdwren/PReNet

openalex publication_date 2019/01/26 · arxiv created 2019/05/16 · arxiv updated 2019/05/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Along with the deraining performance improvement of deep networks, their structures and learning become more and more complicated and diverse, making it difficult to analyze the contribution of various network modules when developing new deraining networks. To handle this issue, this paper provides a better and simpler baseline deraining network by considering network architecture, input and output, and loss functions. Specifically, by repeatedly unfolding a shallow ResNet, progressive ResNet (PRN) is proposed to take advantage of recursive computation. A recurrent layer is further introduced to exploit the dependencies of deep features across stages, forming our progressive recurrent network (PReNet). Furthermore, intra-stage recursive computation of ResNet can be adopted in PRN and PReNet to notably reduce network parameters with graceful degradation in deraining performance. For network input and output, we take both stage-wise result and original rainy image as input to each ResNet and finally output the prediction of residual image. As for loss functions, single MSE or negative SSIM losses are sufficient to train PRN and PReNet. Experiments show that PRN and PReNet perform favorably on both synthetic and real rainy images. Considering its simplicity, efficiency and effectiveness, our models are expected to serve as a suitable baseline in future deraining research. The source codes are available at https://github.com/csdwren/PReNet.

Citations

• Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
• Long Short-Term Memory
• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• Lightweight Pyramid Networks for Image Deraining

Cited by

• Unified Diffusion Transformer for High-fidelity Text-Aware Image Restoration
• TIDE: Two-Stage Inverse Degradation Estimation with Guided Prior Disentanglement for Underwater Image Restoration
• Exploring the Potentials of Spiking Neural Networks for Image Deraining
• SD-PSFNet: Sequential and Dynamic Point Spread Function Network for Image Deraining
• Semantics and Content Matter: Towards Multi-Prior Hierarchical Mamba for Image Deraining
• Rethinking Nighttime Image Deraining via Learnable Color Space Transformation
• WaMaIR: Image Restoration via Multiscale Wavelet Convolutions and Mamba-based Channel Modeling with Texture Enhancement
• Clear Roads, Clear Vision: Advancements in Multi-Weather Restoration for Smart Transportation
• DeRainMamba: A Frequency-Aware State Space Model with Detail Enhancement for Image Deraining
• PRISM: Progressive Rain removal with Integrated State-space Modeling
• Unleashing the Potential of the Semantic Latent Space in Diffusion Models for Image Dehazing
• Degradation-Aware All-in-One Image Restoration via Latent Prior Encoding
• NDLPNet: A Location-Aware Nighttime Deraining Network and a Real-World Benchmark Dataset
• WeatherBench: A Real-World Benchmark Dataset for All-in-One Adverse Weather Image Restoration
• Seeing through Unclear Glass: Occlusion Removal with One Shot
• Single Image Rain Streak Removal Using Harris Corner Loss and R-CBAM Network
• Continual Learning-Based Unified Model for Unpaired Image Restoration Tasks
• DiNAT-IR: Exploring Dilated Neighborhood Attention for High-Quality Image Restoration
• Global Modeling Matters: A Fast, Lightweight and Effective Baseline for Efficient Image Restoration
• EAMamba: Efficient All-Around Vision State Space Model for Image Restoration
• Elucidating and Endowing the Diffusion Training Paradigm for General Image Restoration
• Frequency-Domain Fusion Transformer for Image Inpainting
• Reversing Flow for Image Restoration
• ControlMambaIR: Conditional Controls with State-Space Model for Image Restoration
• Towards Better De-raining Generalization via Rainy Characteristics Memorization and Replay
• CLIP-driven rain perception: Adaptive deraining with pattern-aware network routing and mask-guided cross-attention
• A Regularization-Guided Equivariant Approach for Image Restoration
• DerainSplat: Feed-Forward Clean 3D Gaussian Splatting from Sparse Rainy Views
• Single-image raindrop removal using concurrent channel-spatial attention and long-short skip connections
• DPMambaIR: All-in-One Image Restoration via Degradation-Aware Prompt State Space Model
• RouteWinFormer: A Route-Window Transformer for Middle-range Attention in Image Restoration
• Cross Paradigm Representation and Alignment Transformer for Image Deraining
• Any Image Restoration via Efficient Spatial-Frequency Degradation Adaptation
• Multiscale Tensor Summation Factorization as a New Neural Network Layer (MTS Layer) for Multidimensional Data Processing

Related

• SAPNet: Segmentation-Aware Progressive Network for Perceptual Contrastive Deraining
• Residual-Guide Feature Fusion Network for Single Image Deraining
• Networks are Slacking Off: Understanding Generalization Problem in Image Deraining
• Deep Image Deraining Via Intrinsic Rainy Image Priors and Multi-scale Auxiliary Decoding
• Physical Model Guided Deep Image Deraining