https://github.com/megengine/nbnet
NBNet: Noise Basis Learning for Image Denoising with Subspace Projection
https://github.com/megengine/nbnet
denoise megengine
Last synced: 10 months ago
JSON representation
NBNet: Noise Basis Learning for Image Denoising with Subspace Projection
- Host: GitHub
- URL: https://github.com/megengine/nbnet
- Owner: MegEngine
- License: apache-2.0
- Created: 2021-03-01T14:37:46.000Z (almost 5 years ago)
- Default Branch: main
- Last Pushed: 2021-09-23T10:50:54.000Z (over 4 years ago)
- Last Synced: 2025-04-12T06:51:45.305Z (10 months ago)
- Topics: denoise, megengine
- Language: Python
- Homepage:
- Size: 1.83 MB
- Stars: 154
- Watchers: 5
- Forks: 23
- Open Issues: 13
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# NBNet: Noise Basis Learning for Image Denoising with Subspace Projection
Code for CVPR21 paper [NBNet](https://arxiv.org/abs/2012.15028).
*The illustration of our key insight:*
## Dependencies
- MegEngine >= 1.3.1 (For DistributedDataParallel)
## Training
### Preparation
```
python prepare_data.py --data_dir yours_sidd_data_path
```
### Begin training:
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
```
### Begin testing:
Download the pretrained checkpoint and use:
```
python test.py -d prepared_data_path -c checkpoint_path
```
The result is **PSNR 39.765**.
## Pretrained model
MegEngine checkpoint for SIDD benchmark can be downloaded via
[Google Drive](https://drive.google.com/file/d/1RPAf9ZJqqq9ePPVTtJRlixX4-h3HJTCc/view?usp=sharing)
or
[GitHub Release](https://github.com/megvii-research/NBNet/releases).