app.py

import os

import cv2
import gradio as gr
import torch
from basicsr.archs.srvgg_arch import SRVGGNetCompact
from gfpgan.utils import GFPGANer
from huggingface_hub import snapshot_download, hf_hub_download
from realesrgan.utils import RealESRGANer

REALESRGAN_REPO_ID = 'leonelhs/realesrgan'
GFPGAN_REPO_ID = 'leonelhs/gfpgan'

os.system("pip freeze")

torch.hub.download_url_to_file(
    'https://upload.wikimedia.org/wikipedia/commons/thumb/a/ab/Abraham_Lincoln_O-77_matte_collodion_print.jpg/1024px-Abraham_Lincoln_O-77_matte_collodion_print.jpg',
    'lincoln.jpg')
torch.hub.download_url_to_file(
    'https://user-images.githubusercontent.com/17445847/187400315-87a90ac9-d231-45d6-b377-38702bd1838f.jpg',
    'AI-generate.jpg')
torch.hub.download_url_to_file(
    'https://user-images.githubusercontent.com/17445847/187400981-8a58f7a4-ef61-42d9-af80-bc6234cef860.jpg',
    'Blake_Lively.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 = hf_hub_download(repo_id=REALESRGAN_REPO_ID, filename='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):
    # 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 > 3500 or w > 3500:
            print('too large size')
            return None, None

        if h < 300:
            img = cv2.resize(img, (w * 2, h * 2), interpolation=cv2.INTER_LANCZOS4)

        face_enhancer = None
        snapshot_folder = snapshot_download(repo_id=GFPGAN_REPO_ID)

        if version == 'v1.2':
            path = os.path.join(snapshot_folder, 'GFPGANv1.2.pth')
            face_enhancer = GFPGANer(
                model_path=path, upscale=2, arch='clean', channel_multiplier=2, bg_upsampler=upsampler)
        elif version == 'v1.3':
            path = os.path.join(snapshot_folder, 'GFPGANv1.3.pth')
            face_enhancer = GFPGANer(
                model_path=path, upscale=2, arch='clean', channel_multiplier=2, bg_upsampler=upsampler)
        elif version == 'v1.4':
            path = os.path.join(snapshot_folder, 'GFPGANv1.4.pth')
            face_enhancer = GFPGANer(
                model_path=path, upscale=2, arch='clean', channel_multiplier=2, bg_upsampler=upsampler)
        elif version == 'RestoreFormer':
            path = os.path.join(snapshot_folder, 'RestoreFormer.pth')
            face_enhancer = GFPGANer(
                model_path=path, upscale=2, arch='RestoreFormer', channel_multiplier=2,
                bg_upsampler=upsampler)

        try:
            _, _, 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
    except Exception as error:
        print('global exception', error)
        return None, None


title = "GFPGAN: Practical Face Restoration Algorithm"
description = r"""Gradio demo for <a href='https://github.com/TencentARC/GFPGAN' target='_blank'><b>GFPGAN: Towards Real-World Blind Face Restoration with Generative Facial Prior</b></a>.<br>
It can be used to restore your **old photos** or improve **AI-generated faces**.<br>
To use it, simply upload your image.<br>
If GFPGAN is helpful, please help to ⭐ the <a href='https://github.com/TencentARC/GFPGAN' target='_blank'>Github Repo</a> and recommend it to your friends 😊
"""
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-<COLOR>.svg)](https://arxiv.org/abs/2101.04061)

If you have any question, please email 📧 `xintao.wang@outlook.com` or `xintaowang@tencent.com`.

<center><img src='https://visitor-badge.glitch.me/badge?page_id=akhaliq_GFPGAN' alt='visitor badge'></center>
<center><img src='https://visitor-badge.glitch.me/badge?page_id=Gradio_Xintao_GFPGAN' alt='visitor badge'></center>
"""
demo = gr.Interface(
    inference, [
        gr.Image(type="filepath", label="Input"),
        # gr.Radio(['v1.2', 'v1.3', 'v1.4', 'RestoreFormer', 'CodeFormer'], type="value", value='v1.4', label='version'),
        gr.Radio(['v1.2', 'v1.3', 'v1.4', 'RestoreFormer'], type="value", value='v1.4', label='version'),
        gr.Number(label="Rescaling factor", value=2),
        # gr.Slider(0, 100, label='Weight, only for CodeFormer. 0 for better quality, 100 for better identity', value=50)
    ], [
        gr.Image(type="numpy", label="Output (The whole image)"),
        gr.File(label="Download the output image")
    ],
    title=title,
    description=description,
    article=article,
    examples=[['AI-generate.jpg', 'v1.4', 2],
              ['lincoln.jpg', 'v1.4', 2],
              ['Blake_Lively.jpg', 'v1.4', 2],
              ['10045.png', 'v1.4', 2]])
demo.queue().launch()