import os
import cv2
import gradio as gr
import torch
from basicsr.archs.srvgg_arch import SRVGGNetCompact
from gfpgan.utils import GFPGANer
from realesrgan.utils import RealESRGANer
os.system("pip freeze")
# download weights
if not os.path.exists('realesr-general-x4v3.pth'):
os.system("wget https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-x4v3.pth -P .")
if not os.path.exists('GFPGANv1.2.pth'):
os.system("wget https://github.com/TencentARC/GFPGAN/releases/download/v1.3.0/GFPGANv1.2.pth -P .")
if not os.path.exists('GFPGANv1.3.pth'):
os.system("wget https://github.com/TencentARC/GFPGAN/releases/download/v1.3.0/GFPGANv1.3.pth -P .")
if not os.path.exists('GFPGANv1.4.pth'):
os.system("wget https://github.com/TencentARC/GFPGAN/releases/download/v1.3.0/GFPGANv1.4.pth -P .")
if not os.path.exists('RestoreFormer.pth'):
os.system("wget https://github.com/TencentARC/GFPGAN/releases/download/v1.3.4/RestoreFormer.pth -P .")
if not os.path.exists('CodeFormer.pth'):
os.system("wget https://github.com/TencentARC/GFPGAN/releases/download/v1.3.4/CodeFormer.pth -P .")
if not os.path.exists('HanamichiSakuragi.jpg'):
torch.hub.download_url_to_file(
'https://haoluobo.com/wp-content/uploads/2023/01/%E6%A8%B1%E6%9C%A8%E8%8A%B1%E9%81%93.jpg',
'HanamichiSakuragi.jpg')
torch.hub.download_url_to_file(
'https://haoluobo.com/wp-content/uploads/2023/01/%E6%9D%8E%E4%B8%96%E6%B0%91.jpg',
'LiShiming.jpg')
torch.hub.download_url_to_file(
'https://haoluobo.com/wp-content/uploads/2023/01/%E4%B9%BE%E9%9A%86.jpg',
'QianLong.jpg')
torch.hub.download_url_to_file(
'https://user-images.githubusercontent.com/17445847/187401133-8a3bf269-5b4d-4432-b2f0-6d26ee1d3307.png',
'10045.png')
# background enhancer with RealESRGAN
model = SRVGGNetCompact(num_in_ch=3, num_out_ch=3, num_feat=64, num_conv=32, upscale=4, act_type='prelu')
model_path = 'realesr-general-x4v3.pth'
half = True if torch.cuda.is_available() else False
upsampler = RealESRGANer(scale=4, model_path=model_path, model=model, tile=0, tile_pad=10, pre_pad=0, half=half)
os.makedirs('output', exist_ok=True)
# def inference(img, version, scale, weight):
def inference(img, version, scale, blur_face):
blur_face = int(blur_face)
if blur_face % 2 != 1:
blur_face += 1
if blur_face < 3:
blur_face = 0
# weight /= 100
print(img, version, scale)
if scale > 4:
scale = 4 # avoid too large scale value
try:
extension = os.path.splitext(os.path.basename(str(img)))[1]
img = cv2.imread(img, cv2.IMREAD_UNCHANGED)
if len(img.shape) == 3 and img.shape[2] == 4:
img_mode = 'RGBA'
elif len(img.shape) == 2: # for gray inputs
img_mode = None
img = cv2.cvtColor(img, cv2.COLOR_GRAY2BGR)
else:
img_mode = None
h, w = img.shape[0:2]
if h < 300:
img = cv2.resize(img, (w * 2, h * 2), interpolation=cv2.INTER_LANCZOS4)
if version == 'v1.2':
face_enhancer = GFPGANer(
model_path='GFPGANv1.2.pth', upscale=2, arch='clean', channel_multiplier=2, bg_upsampler=upsampler)
elif version == 'v1.3':
face_enhancer = GFPGANer(
model_path='GFPGANv1.3.pth', upscale=2, arch='clean', channel_multiplier=2, bg_upsampler=upsampler)
elif version == 'v1.4':
face_enhancer = GFPGANer(
model_path='GFPGANv1.4.pth', upscale=2, arch='clean', channel_multiplier=2, bg_upsampler=upsampler)
elif version == 'RestoreFormer':
face_enhancer = GFPGANer(
model_path='RestoreFormer.pth', upscale=2, arch='RestoreFormer', channel_multiplier=2, bg_upsampler=upsampler)
# elif version == 'CodeFormer':
# face_enhancer = GFPGANer(
# model_path='CodeFormer.pth', upscale=2, arch='CodeFormer', channel_multiplier=2, bg_upsampler=upsampler)
try:
# _, _, output = face_enhancer.enhance(img, has_aligned=False, only_center_face=False, paste_back=True, weight=weight)
face_helper = face_enhancer.face_helper
align_warp_face = face_helper.align_warp_face
def new_align_warp_face(*args, **kwargs):
align_warp_face(*args, **kwargs) # save_cropped_path
face_helper.org_cropped_faces = face_helper.cropped_faces
if blur_face >= 3:
face_helper.cropped_faces = [cv2.GaussianBlur(e, (blur_face, blur_face), 0) for e in face_helper.cropped_faces]
print("find face count:", len(face_helper.cropped_faces))
face_helper.align_warp_face = new_align_warp_face
_, _, output = face_enhancer.enhance(img, has_aligned=False, only_center_face=False, paste_back=True)
except RuntimeError as error:
print('Error', error)
try:
if scale != 2:
interpolation = cv2.INTER_AREA if scale < 2 else cv2.INTER_LANCZOS4
h, w = img.shape[0:2]
output = cv2.resize(output, (int(w * scale / 2), int(h * scale / 2)), interpolation=interpolation)
except Exception as error:
print('wrong scale input.', error)
if img_mode == 'RGBA': # RGBA images should be saved in png format
extension = 'png'
else:
extension = 'jpg'
save_path = f'output/out.{extension}'
cv2.imwrite(save_path, output)
output = cv2.cvtColor(output, cv2.COLOR_BGR2RGB)
return (
output,
save_path,
[cv2.cvtColor(e, cv2.COLOR_BGR2RGB) for e in face_enhancer.face_helper.org_cropped_faces],
[cv2.cvtColor(e, cv2.COLOR_BGR2RGB) for e in face_enhancer.face_helper.restored_faces]
)
except Exception as error:
print('global exception', error)
return None, None
title = "GFPGAN: Practical Face Restoration Algorithm"
description = r"""Gradio demo for GFPGAN: Towards Real-World Blind Face Restoration with Generative Facial Prior.
It can be used to restore your **old photos** or improve **AI-generated faces**.
To use it, simply upload your image.
If GFPGAN is helpful, please help to ⭐ the Github Repo and recommend it to your friends 😊
This demo was forked by [vicalloy](https://github.com/vicalloy), add `face blur` param to optimize painting face enhance.
"""
article = r"""
[![download](https://img.shields.io/github/downloads/TencentARC/GFPGAN/total.svg)](https://github.com/TencentARC/GFPGAN/releases)
[![GitHub Stars](https://img.shields.io/github/stars/TencentARC/GFPGAN?style=social)](https://github.com/TencentARC/GFPGAN)
[![arXiv](https://img.shields.io/badge/arXiv-Paper-.svg)](https://arxiv.org/abs/2101.04061)
If you have any question, please email 📧 `xintao.wang@outlook.com` or `xintaowang@tencent.com`.
"""
with gr.Blocks() as demo:
gr.Markdown("%s
" % title)
gr.Markdown(description)
with gr.Row(equal_height=False):
with gr.Column():
file_path = gr.components.Image(type="filepath", label="Input")
version = gr.components.Radio(['v1.2', 'v1.3', 'v1.4', 'RestoreFormer'], type="value", value='v1.4', label='version')
rescaling_factor = gr.components.Number(label="Rescaling factor", value=2)
blur_face = gr.components.Number(label="Blur face", value=25)
submit = gr.Button("Submit")
with gr.Column():
output_img = gr.components.Image(type="numpy", label="Output (The whole image)")
download = gr.components.File(label="Download the output image")
with gr.Row():
with gr.Column():
input_faces = gr.Gallery(label="Input faces").style(height="auto")
with gr.Column():
output_faces = gr.Gallery(label="Output faces").style(height="auto")
gr.Examples([['HanamichiSakuragi.jpg', 'v1.4', 2, 31], ['LiShiming.jpg', 'v1.4', 2, 3], ['QianLong.jpg', 'v1.4', 2, 3],
['10045.png', 'v1.4', 2, 0]], [file_path, version, rescaling_factor, blur_face])
gr.Markdown(article)
submit.click(
inference,
inputs=[file_path, version, rescaling_factor, blur_face],
outputs=[output_img, download, input_faces, output_faces]
)
demo.queue(concurrency_count=4)
demo.launch()