Abstract:Deraining is a significant and fundamental computer vision task, aiming to remove the rain streaks and accumulations in an image or video. Existing deraining methods usually make heuristic assumptions of the rain model, which compels them to employ complex optimization or iterative refinement for high recovery quality. However, this leads to time-consuming methods and affects the effectiveness of addressing rain patterns, deviating from the assumptions. This paper proposes a simple yet efficient deraining method by formulating deraining as a predictive filtering problem without complex rain model assumptions. Specifically, we identify spatially-variant predictive filtering (SPFilt) that adaptively predicts proper kernels via a deep network to filter different individual pixels. Since the filtering can be implemented via well-accelerated convolution, our method can be significantly efficient. We further propose the EfDeRain+ that contains three main contributions to address residual rain traces, multi-scale, and diverse rain patterns without harming efficiency. First , we propose the uncertainty-aware cascaded predictive filtering (UC-PFilt) that can identify the difficulties of reconstructing clean pixels via predicted kernels and remove the residual rain traces effectively. Second , we design the weight-sharing multi-scale dilated filtering (WS-MS-DFilt) to handle multi-scale rain streaks without harming the efficiency. Third , to eliminate the gap across diverse rain patterns, we propose a novel data augmentation method ( i.e ., RainMix ) to train our deep models. By combining all contributions with sophisticated analysis on different variants, our final method outperforms baseline methods on six single-image deraining datasets and one video-deraining dataset in terms of both recovery quality and speed. In particular, EfDeRain+ can derain within about 6.3 ms on a image and is over 74 times faster than the top baseline method with even better recovery quality. We release code in https://github.com/tsingqguo/efficientderainplus.

Sparse Sampling Transformer with Uncertainty-Driven Ranking for Unified Removal of Raindrops and Rain Streaks

Multi-patch de-raindrop Transformer for UAV images

Uncertainty-Aware Cascaded Dilation Filtering for High-Efficiency Deraining

Mutual Channel Prior Guided Dual-Domain Transformer for Single Image Raindrop Removal

Image De-Raining Transformer

Detail-recovery Image Deraining via Dual Sample-augmented Contrastive Learning

Spatial Attentive Single-Image Deraining With a High Quality Real Rain Dataset

EfficientDeRain+: Learning Uncertainty-Aware Filtering via RainMix Augmentation for High-Efficiency Deraining

DR-DiT: Image Deraining Using Diffusion Model with Transformer

Not All Areas Are Equal: A Novel Separation‐Restoration‐Fusion Network for Image Raindrop Removal

Not Just Streaks: Towards Ground Truth for Single Image Deraining

MLTDNet: an efficient multi-level transformer network for single image deraining

A deraining with detail-recovery network via context aggregation

Towards an Effective and Efficient Transformer for Rain-by-snow Weather Removal

RainFormer: a Pyramid Transformer for Single Image Deraining

Dbswin: Transformer Based Dual Branch Network for Single Image Deraining

Context-Aware Transformer for Single Image Rain Streaks Removal

Learning A Sparse Transformer Network for Effective Image Deraining

Hybrid CNN-Transformer Feature Fusion for Single Image Deraining

DRD-Net: Detail-recovery Image Deraining via Context Aggregation Networks