This is a PyTorch/GPU implementation of our Information Fusion 2022 paper: Rethinking multi-exposure image fusion with extreme and diverse exposure levels: A robust framework based on Fourier transform and contrastive learning.
- Please refer to the training code for model training.
- Please refer to the testing code for image fusion.
We construct two new MEF benchmark test sets, eMEFB and rMEFB, which can be used to evaluate MEF algorithms in fusing image pairs with extreme exposure levels and image pairs under diverse combinations of random exposure levels, respectively. The eMEFB dataset and the rMEFB dataset can be downloaded from this link.
For more details, please refer to the paper and this repo.
If this work is helpful to you, please cite it as:
@article{qu2023rethinking,
title={Rethinking multi-exposure image fusion with extreme and diverse exposure levels: A robust framework based on Fourier transform and contrastive learning},
author={Qu, Linhao and Liu, Shaolei and Wang, Manning and Song, Zhijian},
journal={Information Fusion},
volume={92},
pages={389--403},
year={2023},
publisher={Elsevier}
}
If you have any question, please email to me [email protected].