Depth-Attentional Features for Single-Image Rain Removal

By Xiaowei Hu, Chi-Wing Fu, Lei Zhu, P. Heng. Computer Vision and Pattern Recognition, 2019

400 citations recorded by the source corpus.

Rain is a common weather phenomenon, where object visibility varies with depth from the camera and objects faraway are visually blocked more by fog than by rain streaks. Existing methods and datasets for rain removal, however, ignore these physical properties, thereby limiting the rain removal efficiency on real photos. In this work, we first analyze the visual effects of rain subject to scene depth and formulate a rain imaging model collectively with rain streaks and fog; by then, we prepare a new dataset called RainCityscapes with rain streaks and fog on real outdoor photos. Furthermore, we design an end-to-end deep neural network, where we train it to learn depth-attentional features via a depth-guided attention mechanism, and regress a residual map to produce the rain-free image output. We performed various experiments to visually and quantitatively compare our method with several state-of-the-art methods to demonstrate its superiority over the others.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/eee324d41e696b743f6f2f9ce0a46d1ac19f08c8

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Learning A Sparse Transformer Network for Effective Image Deraining (2023, neutral): “To better represent the rain distribution, several studies take rain characteristics such as rain direction [27], density [61], veiling effect [15] into account, and optimize the network structure via recursive computation [19, 23, 36] or transfer mechanism [16, 49, 54, 55].”
RobustNet: Improving Domain Generalization in Urban-Scene Segmentation via Instance Selective Whitening (2021, methodology): “Segmentation results on unseen domains ( i.e., BDD-100K [63] and RainCityscapes [20]) with the models trained on Cityscapes [9].”
Cooperative Semantic Segmentation and Image Restoration in Adverse Environmental Conditions (2019, background): “This dataset, named ‘RainCityscapes’, has 9,432 training images and 1,188 test images.”
Defocus Blur Detection via Depth Distillation (2020, methodology): “Some methods exploit the knowledge of depth in related tasks, such as depth-assisted view synthesis [4] and depth attentional features for deraining [12].”