NLEDN

Non-locally Enhanced Encoder-Decoder Network for Single Image De-raining

The testing code of Non-locally Enhanced Encoder-Decoder Network for Single Image De-raining.

Citation

@article{li2018non-locally,
    title={Non-locally Enhanced Encoder-Decoder Network for Single Image De-raining},
    author={Li, Guanbin and He, Xiang and Zhang, Wei and Chang, Huiyou and Dong, Le and Lin, Liang},
    journal={acm multimedia},
    year={2018}
}

Visualization

Environment

• python 3.6
• pytorch 0.4.0

Download Link (here)

• Four testing sets.
• De-rained results.
• Four model weights.

Running Steps

1. Download four testing sets (Rain100L, Rain100H, DDN, DID-MDN).
2. Choose the type of dataset in config.py.

   data_type_list = ['Rain100L', 'Rain100H', 'DDN', 'DID']
   data_type = data_type_list[2]
   

3. Run test.py.

   python test.py