Learning-to-See-in-the-Dark

Pytorch implementation of Learning to See in the Dark (CVPR 2018)

Developed a pipeline for processing low-light images, based on end-to-end training of a fully convolutional network. The network operates directly on raw sensor data and the original paper referenced can be found here.

Data

For this implementation, only images from the Sony camera has been used. The author has provided this dataset here. The resolution is 4240x2832 for each raw image. I process these images and convert each of them into .npy file for reducing the training time.

Usage

Producing image

python main.py eval --dark_image </path/to/raw/image> --model_path </path/to/saved/model> --output_image </path/to/output/image> --cuda 1

• --dark_image path of the raw image (ARW) you want to evaluate
• --model saved model to be used
• --scale_percent scale down the input bayer image by this percent (helpful for low memory GPUs), default is 100
• --output_image path for saving the output image
• --cuda set it to 1 for running on GPU, 0 for CPU. (default is 1)

Train model

python main.py train --train_txt_file </path/to/train/text/file> --patch_size 512 --epochs 4000 --save_model_dir </path/to/save-model/folder> --cuda 1

• --train_txt_file path to the txt file corresponding to the training data, default is ./Sony/Sony_train_list.txt
• --patch_size patch size used while training, default is 512
• --epochs number of training epochs, default is 4000
• --save_model_dir path to folder where trained model will be saved, default is ./checkpoint_result/
• --cuda set it to 1 for running on GPU, 0 for CPU. (default is 1)

Refer to ./main.py for other command line arguments.

More Info

Learn about U-Net architecture here. Know about Rawpy.