Ensemble single image deraining network via progressive structural boosting constraints

作者：

Highlights：

• The ensemble network has capacity to solve rain removal of multiple types.

• Struct-absolute loss benefits for embedding boosting strategy into network training.

• The ensemble network achieves new SOTA performance on public deraining datasets.

摘要

•The ensemble network has capacity to solve rain removal of multiple types.•Struct-absolute loss benefits for embedding boosting strategy into network training.•The ensemble network achieves new SOTA performance on public deraining datasets.

论文关键词：Single image derain,Boosting,Ensemble derain network,Rain removal

论文评审过程：Received 14 October 2020, Revised 23 April 2021, Accepted 23 August 2021, Available online 4 September 2021, Version of Record 9 September 2021.

论文官网地址：https://doi.org/10.1016/j.image.2021.116460