Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
You can not select more than 25 topics
Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
tiantian132
32c03e1047
|
1 year ago | |
|---|---|---|
| .. | ||
| Datasets | 1 year ago | |
| pretrained_models | 1 year ago | |
| utils | 1 year ago | |
| MPRNet.py | 1 year ago | |
| README.md | 1 year ago | |
| config.py | 1 year ago | |
| data_RGB.py | 1 year ago | |
| dataset_RGB.py | 1 year ago | |
| evaluate_GOPRO_HIDE.m | 1 year ago | |
| evaluate_RealBlur.py | 1 year ago | |
| losses.py | 1 year ago | |
| test.py | 1 year ago | |
| train.py | 1 year ago | |
| training.yml | 1 year ago | |
README.md
Training
-
Download the Datasets
-
Train the model with default arguments by running
python train.py
Evaluation
Download the model and place it in ./pretrained_models/
Testing on GoPro dataset
- Download images of GoPro and place them in
./Datasets/GoPro/test/ - Run
python test.py --dataset GoPro
Testing on HIDE dataset
- Download images of HIDE and place them in
./Datasets/HIDE/test/ - Run
python test.py --dataset HIDE
Testing on RealBlur-J dataset
- Download images of RealBlur-J and place them in
./Datasets/RealBlur_J/test/ - Run
python test.py --dataset RealBlur_J
Testing on RealBlur-R dataset
- Download images of RealBlur-R and place them in
./Datasets/RealBlur_R/test/ - Run
python test.py --dataset RealBlur_R
To reproduce PSNR/SSIM scores of the paper on GoPro and HIDE datasets, run this MATLAB script
evaluate_GOPRO_HIDE.m
To reproduce PSNR/SSIM scores of the paper on RealBlur dataset, run
evaluate_RealBlur.py
