Single Image De Raining Using Deep Decomposition Composition Network

Abstract

Under rainy conditions the impact of rain streaks on images and video is often undesirable. The effects of rain can also severely affect the performance of outdoor vision system. The quality of the image is degraded by rain streaks. Hence it is necessary to remove rain streaks from single image which is a challenging problem. Towards fixing this problem the deep decomposition-composition network is proposed. This paper designs a novel multi-task leaning architecture in an end-to-end manner to reduce the mapping range from input to output and boost the performance. Concretely, a decomposition net is built to split rain images into clean background and rain layers. Different from previous architectures, this model consists of, besides a component representing the desired clean image, an extra component for the rain layer. During the training phase, further employ a composition structure to reproduce the input by the separated clean image and rain information for improving the quality of decomposition. Furthermore, this design is also applicable to other layer decomposition tasks like dust removal. More importantly, this method only requires about 50ms, significantly faster than the competitors, to process a testing image in VGA resolution on a GTX 1080 GPU, making it attractive for practical use