We propose a plug-and-play (PnP) method, which uses deep-learning-based denoisers as regularization priors for spectral snapshot compressive imaging (SCI). Our method is efficient in terms of reconstruction quality and speed trade-off, and flexible to be ready-to-use for different compressive coding mechanisms. We demonstrate the efficiency and flexibility in both simulations and five different spectral SCI systems and show that the proposed deep PnP prior could achieve state-of-the-art results with a simple plug-in based on the optimization framework. This paves the way for capturing and recovering multi- or hyper-spectral information in one snapshot, which might inspire intriguing applications in remote sensing, biomedical science and material science.
@article{Zheng20PnPCASSI,
author = {Zheng, Siming and Liu, Yang and Meng, Ziyi and Qiao, Mu and Tong, Zhishen and Yang, Xiaoyu and Han, Shensheng and Yuan, Xin},
title = {Deep Plug-and-Play Priors for Spectral Snapshot Compressive Imaging},
journal = {Photonics Research},
doi = {10.1364/PRJ.411745},
volume = {9},
number = {2},
pages = {B18--B29},
year = {2021},
url = {https://doi.org/10.1364/PRJ.411745},
publisher = {Optical Society of America},
type = {Journal Article}
}