Code for CVPR21 paper NBNet.
The illustration of our key insight:
- MegEngine >= 1.3.1 (For DistributedDataParallel)
python prepare_data.py --data_dir yours_sidd_data_path
For SIDD benchmark, use:
python train_mge.py -d prepared_data_path -n num_gpus
For DnD benchmark, we use MixUp additionally:
python train_mge.py -d prepared_data_path -n num_gpus --dnd
Download the pretrained checkpoint and use:
python test.py -d prepared_data_path -c checkpoint_path
The result is PSNR 39.765.
MegEngine checkpoint for SIDD benchmark can be downloaded via Google Drive or GitHub Release.