Face-Super-Resolution

Face super resolution based on ESRGAN (https://github.com/xinntao/BasicSR)

Results

INPUT & AFTER-SR & GROUND TRUTH

Usage

Train

Run python gen_lr_imgs.py to get the face imgs with low resolution and pool qualities
Set the dir in train.py

hr_path: The path list of imgs with high resolution.

lr_path: The path of imgs with low resolution.

Run python train.py

Pretrained models

Dlib alignment shape_predictor_68_face_landmarks.dat

(https://pan.baidu.com/s/19Y-AYnXs6ubIh4vlkyvqbQ)

(https://drive.google.com/open?id=1u3h3nX5f_w-HJV8Nd1zwqc3uTnVja5Ol)

Generator weights

90000_G.pth

(https://pan.baidu.com/s/14ITkNz_t0E7hRv0-tTAjhA)

(https://drive.google.com/open?id=1CZkLZPtbJepgksCM93MvsY7NgqnEZSvk)

90000_G.pth (The last activation in G is linear, clearer)

90000_D.pth

(https://pan.baidu.com/s/1-gRy1xw5h95_ie0NfBbDlw password:6him)

(https://drive.google.com/file/d/1zX9dbCu9lFu_SvRCvPIWnwJUUGUrZUhk/view?usp=sharing)

Maybe you need it to finetune the model?

200000_G.pth

(https://pan.baidu.com/s/1Osge_4JjPyvG5Xfnbe9KVA)

(https://drive.google.com/open?id=1B6BQu5Qk8eIu8MGTWJHnJxaxY1zCqQEt)

200000_G.pth (The last activation in G is tanh)

Test

Download 'shape_predictor_68_face_landmarks.dat' and '90000_G.pth'
Set 'pretrain_model_G' in test.py
RUN python test.py