Medical_Image

Results

• Results on patches
• Results on patients

Performance Autoencoder approach:

Performance CNN approach:

1. Create Environment

• Make Conda Environment

Ficar com crear el environment desde el yml
conda create -n x python=3.7
conda activate x

 

2. Prepare Dataset

Download the following datasets:

LOL-v1 Baidu Disk (code: cyh2), Google Drive

 

3. Testing

Download our models from Baidu Disk (code: cyh2) or Google Drive. Put them in folder pretrained_weights

# Autoencoder
python3 Enhancement/test_from_dataset.py --opt Options/RetinexFormer_LOL_v1.yml --weights pretrained_weights/LOL_v1.pth --dataset LOL_v1

# CNN
python3 Enhancement/test_from_dataset.py --opt Options/RetinexFormer_LOL_v2_real.yml --weights pretrained_weights/LOL_v2_real.pth --dataset LOL_v2_real

• Evaluating the Params and FLOPS of models

 

4. Training

Feel free to check our training logs from Baidu Disk (code: cyh2) or Google Drive

# Autoencoder
python3 basicsr/train.py --opt Options/RetinexFormer_LOL_v1.yml

# CNN
python3 basicsr/train.py --opt Options/RetinexFormer_LOL_v2_real.yml