This repository contains the python code for the paper Wu, Z., Yang, C., Su, X., & Yuan, X. (2023). Adaptive Deep PnP Algorithm for Video Snapshot Compressive Imaging. International Journal of Computer Vision, https://doi.org/10.1007/s11263-023-01777-y..
numpy==1.21.2
pytorch==1.8.1
scipy==1.7.1
scikit-image==0.18.1
h5py==2.10.0
tqdm==4.61.2
tensorboardx==2.2
Please add simulation middle scale color data from PnP_SCI to ./dataset.
Our results are saved in OneDrive Link.
Run ADMM_TV_Warm_Start_save.py
to save the TV prior initialized results.
[OPTION] Or just add the TV prior initialized results (_Admm_tv_xxx_bayer8.mat
) from OneDrive Link to ./results/savedmat/
.
Run twoStageADMM_Online_FFD_WARM.py
or twoStageADMM_Online_FastDVD_WARM.py
to test the algorithm after loading results initialized with TV prior as warn start.
directory | description |
---|---|
dataset |
data original |
packages and models |
algorithms pluged into PnP framework |
model_zoo |
pre-trained model data |
dataset |
data used for reconstruction (simulated or real data) |
results |
results of reconstruction (after reconstruction) |
utils |
utility functions |
@article{wu_adaptive_2023,
title = {Adaptive {Deep} {PnP} {Algorithm} for {Video} {Snapshot} {Compressive} {Imaging}},
issn = {1573-1405},
url = {https://doi.org/10.1007/s11263-023-01777-y},
doi = {10.1007/s11263-023-01777-y},
journal = {International Journal of Computer Vision},
author = {Wu, Zongliang and Yang, Chengshuai and Su, Xiongfei and Yuan, Xin},
month = mar,
year = {2023},
}