Awesome

Spatial-Frequency Mutual Learning for Face Super-Resolution

Pretrained Model

BaiDu passward:en47, Google Drive

Requirement

Pytorch 1.11.0 Cuda 11.4

Train

Train PSNR-oriented model:

python main_two.py --data_path data_path  --writer_name mynet --scale scale --model MYNET

Train GAN-based model:

python main_GAN.py --load pretrained_PSNR_model --model MYNET --scale 8 --data_path data_path  --writer_name mynetgan

Test

python test.py --model MYNET --load pretrained_model_path  --data_path data_path --save_name name

Test dataset

BaiDu passward:v2uf

Citation

@InProceedings{Wang_2023_CVPR,
    author    = {Wang, Chenyang and Jiang, Junjun and Zhong, Zhiwei and Liu, Xianming},
    title     = {Spatial-Frequency Mutual Learning for Face Super-Resolution},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2023},
    pages     = {22356-22366}
}