Rethinking Nighttime Image Deraining via Learnable Color Space Transformation

Qiyuan Guan (Dalian Polytechnic University) · Xiang Chen (Nanjing University of Science and Technology) · Guiyue Jin (Dalian Polytechnic University) · Jiyu Jin (Dalian Polytechnic University) · Shumin Fan (Dalian Polytechnic University) · Tianyu Song (Dalian Martime University) · Jinshan Pan (Nanjing University of Science and Technology)

color space transformation networkcomplex scenarioscst-netdataset effectivenessfeature learninghigh-quality benchmarkhq-nightrainillumination informationimplicit illumination guidancelearnable color space convertermodel robustnessnighttime image derainingnighttime scenariosrain removalrgb color spacey channel

Compared to daytime image deraining, nighttime image deraining poses significant challenges due to inherent complexities of nighttime scenarios and the lack of high-quality datasets that accurately represent the coupling effect between rain and illumination. In this paper, we rethink the task of nighttime image deraining and contribute a new high-quality benchmark, HQ-NightRain, which offers higher harmony and realism compared to existing datasets. In addition, we develop an effective Color Space Transformation Network (CST-Net) for better removing complex rain from nighttime scenes. Specifically, we propose a learnable color space converter (CSC) to better facilitate rain removal in the Y channel, as nighttime rain is more pronounced in the Y channel compared to the RGB color space. To capture illumination information for guiding nighttime deraining, implicit illumination guidance is introduced enabling the learned features to improve the model's robustness in complex scenarios. Extensive experiments show the value of our dataset and the effectiveness of our method. The source code and datasets are available at https://github.com/guanqiyuan/CST-Net.