MSBDN-DFF-master
所属分类：图形图像处理
开发工具：Python
文件大小：219KB
下载次数：1
上传日期：2020-10-07 21:41:52
上 传 者：
Sherry_Y
说明： 论文《Multi-Scale Boosted Dehazing Network with Dense Feature Fusion》源码
(Multi-Scale Boosted Dehazing Network with Dense Feature Fusion)
文件列表:
HDF5 (0, 2020-08-08)
HDF5\dehaze_RESIDE_HDF5.m (3938, 2020-08-08)
HDF5\dehaze_RESIDE_HDF5.m~ (3938, 2020-08-08)
HDF5\modcrop.m (294, 2020-08-08)
HDF5\store2hdf5_2out.m (3627, 2020-08-08)
datasets (0, 2020-08-08)
datasets\__pycache__ (0, 2020-08-08)
datasets\__pycache__\dataset_hf5.cpython-36.pyc (9459, 2020-08-08)
datasets\__pycache__\dataset_hf5.cpython-37.pyc (3344, 2020-08-08)
datasets\dataset_hf5.py (3364, 2020-08-08)
networks (0, 2020-08-08)
networks\MSBDN-DFF-v1-1.py (6462, 2020-08-08)
networks\__pycache__ (0, 2020-08-08)
networks\__pycache__\Dense_UNet_v0.cpython-36.pyc (4843, 2020-08-08)
networks\__pycache__\Dense_UNet_v1.cpython-36.pyc (5360, 2020-08-08)
networks\__pycache__\Dense_UNet_v2.cpython-36.pyc (5385, 2020-08-08)
networks\__pycache__\Dense_UNet_v3.cpython-36.pyc (5353, 2020-08-08)
networks\__pycache__\Dense_UNet_v5-1.cpython-36.pyc (5433, 2020-08-08)
networks\__pycache__\Dense_UNet_v5.cpython-36.pyc (5436, 2020-08-08)
networks\__pycache__\Dense_UNet_v7.cpython-36.pyc (5431, 2020-08-08)
networks\__pycache__\DuRN_US.cpython-36.pyc (4942, 2020-08-08)
networks\__pycache__\GCANet.cpython-36.pyc (4385, 2020-08-08)
networks\__pycache__\GCANet.cpython-37.pyc (4268, 2020-08-08)
networks\__pycache__\MSBDN-DFF-v1-1-M.cpython-36.pyc (6600, 2020-08-08)
networks\__pycache__\MSBDN-DFF-v1-1-S.cpython-36.pyc (6191, 2020-08-08)
networks\__pycache__\MSBDN-DFF-v1-1.cpython-36.pyc (6598, 2020-08-08)
networks\__pycache__\MSBDN-v1-1.cpython-36.pyc (6068, 2020-08-08)
networks\__pycache__\MSBDN-v1.cpython-36.pyc (6332, 2020-08-08)
networks\__pycache__\MSBDN.cpython-36.pyc (11299, 2020-08-08)
networks\__pycache__\MSBDN_S.cpython-36.pyc (6568, 2020-08-08)
networks\__pycache__\MSBDN_S1.cpython-36.pyc (6164, 2020-08-08)
networks\__pycache__\MSBDN_rcan_v1.cpython-36.pyc (7401, 2020-08-08)
networks\__pycache__\MSBDN_twcing.cpython-36.pyc (6341, 2020-08-08)
networks\__pycache__\MSBDN_v1.cpython-36.pyc (6585, 2020-08-08)
networks\__pycache__\N_modules.cpython-36.pyc (3223, 2020-08-08)
networks\__pycache__\PFFNet.cpython-36.pyc (5059, 2020-08-08)
networks\__pycache__\PFFNet.cpython-37.pyc (4930, 2020-08-08)
networks\__pycache__\PFFNet_DFF.cpython-36.pyc (6051, 2020-08-08)
... ...
# MSBDN-DFF
The source code of CVPR 2020 paper **"Multi-Scale Boosted Dehazing Network with Dense Feature Fusion"** by [Hang Dong](https://sites.google.com/view/hdong/%E9%A6%96%E9%A1%B5), [Jinshan Pan](https://jspan.github.io/), [Zhe Hu](https://zjuela.github.io/), Xiang Lei, [Xinyi Zhang](http://xinyizhang.tech), Fei Wang, [Ming-Hsuan Yang](http://faculty.ucmerced.edu/mhyang/)
## Dependencies
* Python 3.6
* PyTorch >= 1.1.0
* torchvision
* numpy
* skimage
* h5py
* MATLAB
## Test
1. Download the [Pretrained model on RESIDE](https://drive.google.com/open?id=1da13IOlJ3FQfH6Duj_u1exmZzgXPaYXe) and
[Test set](https://drive.google.com/open?id=1qZlnJN4ybjunc2BGh6kjOUfFdVxuNS-P) to ``MSBDN-DFF/models`` and ``MSBDN-DFF/``folder, respectively.
2. Run the ``MSBDN-DFF/test.py`` with cuda on command line:
```bash
MSBDN-DFF/$python test.py --checkpoint path_to_pretrained_model
```
3. The dehazed images will be saved in the directory of the test set.
## Train
We find the choices of training images play an important role during the training stage, so we offer the training set of HDF5 format:
[Baidu Yun](https://pan.baidu.com/s/1NqAaec3MFwFU9ZM2lfR_4w) (code:v8ku)
You can use the DataSet_HDF5() in ./datasets/dataset_hf5.py to load these HDF5 files.
## Citation
If you use these models in your research, please cite:
@conference{MSBDN-DFF,
author = {Hang, Dong and Jinshan, Pan and Zhe, Hu and Xiang, Lei and Fei, Wang and Ming-Hsuan, Yang},
title = {Multi-Scale Boosted Dehazing Network with Dense Feature Fusion},
booktitle = {CVPR},
year = {2020}
}
近期下载者：
相关文件：
收藏者：