Rain O'er Me: Synthesizing real rain to derain with data distillation
This work addresses the challenge of realistic rain removal in computer vision, which is incremental as it builds on existing deraining methods by avoiding synthetic data limitations.
The paper tackles the problem of removing rain from images without relying on synthetic rain software by using a two-stage data distillation method that pairs rainy images with coarsely derained versions and clean images with generated rainy pairs, resulting in improved deraining by leveraging high-resolution structure from clean images.
We present a supervised technique for learning to remove rain from images without using synthetic rain software. The method is based on a two-stage data distillation approach: 1) A rainy image is first paired with a coarsely derained version using on a simple filtering technique ("rain-to-clean"). 2) Then a clean image is randomly matched with the rainy soft-labeled pair. Through a shared deep neural network, the rain that is removed from the first image is then added to the clean image to generate a second pair ("clean-to-rain"). The neural network simultaneously learns to map both images such that high resolution structure in the clean images can inform the deraining of the rainy images. Demonstrations show that this approach can address those visual characteristics of rain not easily synthesized by software in the usual way.