Recent advances in image deraining have focused on training powerful models
on mixed multiple datasets comprising diverse rain types and backgrounds.
However, this approach tends to overlook the inherent differences among rainy
images, leading to suboptimal results. To overcome this li