in Image Restoration Inverse Problem Computer Vision ~ read.

Semi-supervised CNN for Single Image Rain Removal(ECCV2018)

https://arxiv.org/pdf/1807.11078.pdf

  1. a network supervised by generated rains

  2. fine tuned on real data

inspired by
Wei, W., Yi, L., Xie, Q., Zhao, Q., Meng, D., Xu, Z.: Should we encode rain streaks in video as deterministic or stochastic? In: 2017 IEEE International Conference
on Computer Vision (ICCV). Volume 00. (Oct. 2018) 2535–2544
Idea of this series of work is also interesting: the data term is learned by a em algorithm and this seems fits better on real data.[they have a serious of work publish in all aspect of image restoration.]

We design an EM algorithm together with a gradient descent strategy to
solve the proposed model. The P-MoG parameters and network parameters
can be optimized by sequence in each iteration. Experiments implemented on
synthetic rainy images and especially real ones substantiate the superiority
of the proposed method as compared with the state-of-the-arts along this
line.