app.py

import gradio as gr
import os.path
import numpy as np
from collections import OrderedDict
import torch
import cv2
from PIL import Image, ImageOps
import utils_image as util
from network_fbcnn import FBCNN as net
import requests
import datetime

print(gr.__version__)
for model_path in ['fbcnn_gray.pth','fbcnn_color.pth']:
    if os.path.exists(model_path):
        print(f'{model_path} exists.')
    else:
        print("Downloading model: ", f'{model_path}')
        url = 'https://github.com/jiaxi-jiang/FBCNN/releases/download/v1.0/{}'.format(os.path.basename(model_path))
        r = requests.get(url, allow_redirects=True)
        open(model_path, 'wb').write(r.content)    

def inference(filepaths, is_gray, res_percentage, input_quality):
    outputs = []
    before_afters = []
    if filepaths is None:
      return [], None
    
    for filepath, *_ in filepaths:
      filename = os.path.basename(filepath)
      print("Processing: ", filename)
      input_img = np.array(Image.open(filepath).convert("RGB"))
      
      print("Datetime: ", datetime.datetime.utcnow())
      input_img_width, input_img_height = Image.fromarray(input_img).size
      print("Img size: ", (input_img_width, input_img_height))
  
      resized_input = Image.fromarray(input_img).resize(
          (
              int(input_img_width * (res_percentage/100)),
              int(input_img_height * (res_percentage/100))
          ), resample = Image.BICUBIC)
      input_img = np.array(resized_input)
      print("Input image resized to: ", resized_input.size)
  
      if is_gray:
          n_channels = 1
          model_name = 'fbcnn_gray.pth'
      else:
          n_channels = 3
          model_name = 'fbcnn_color.pth'
      nc = [64,128,256,512]
      nb = 4
  
      input_quality = 100 - input_quality
  
      model_path = model_name
  
      device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
      print("Device: ", device)
  
      print(f'Loading model from {model_path}')
      
      model = net(in_nc=n_channels, out_nc=n_channels, nc=nc, nb=nb, act_mode='R')
      print("#model.load_state_dict(torch.load(model_path), strict=True)")
      model.load_state_dict(torch.load(model_path), strict=True)
      print("#model.eval()")
      model.eval()
      print("#for k, v in model.named_parameters()")
      for k, v in model.named_parameters():
          v.requires_grad = False
      print("#model.to(device)")
      model = model.to(device)
      print("Model loaded.")
  
      test_results = OrderedDict()
      test_results['psnr'] = []
      test_results['ssim'] = []
      test_results['psnrb'] = []
  
      print("#if n_channels")
      if n_channels == 1:
          open_cv_image = Image.fromarray(input_img)
          open_cv_image = ImageOps.grayscale(open_cv_image)
          open_cv_image = np.array(open_cv_image)
          img = np.expand_dims(open_cv_image, axis=2)
      elif n_channels == 3:
          open_cv_image = np.array(input_img)
          if open_cv_image.ndim == 2:
              open_cv_image = cv2.cvtColor(open_cv_image, cv2.COLOR_GRAY2RGB)
          else:
              open_cv_image = cv2.cvtColor(open_cv_image, cv2.COLOR_BGR2RGB)
  
      print("#util.uint2tensor4(open_cv_image)")
      img_L = util.uint2tensor4(open_cv_image)
      
      print("#img_L.to(device)")
      img_L = img_L.to(device)
  
      print("#model(img_L)")
      img_E, QF = model(img_L)
      print("#util.tensor2single(img_E)")
      img_E = util.tensor2single(img_E)
      print("#util.single2uint(img_E)")
      img_E = util.single2uint(img_E)
      
      print("#torch.tensor([[1-input_quality/100]]).cuda() || torch.tensor([[1-input_quality/100]])")
      qf_input = torch.tensor([[1-input_quality/100]]).cuda() if device == torch.device('cuda') else torch.tensor([[1-input_quality/100]])
      print("#util.single2uint(img_E)")
      img_E, QF = model(img_L, qf_input)  
  
      print("#util.tensor2single(img_E)")
      img_E = util.tensor2single(img_E)
      print("#util.single2uint(img_E)")
      img_E = util.single2uint(img_E)
  
      if img_E.ndim == 3:
          img_E = img_E[:, :, [2, 1, 0]]
  
      print("--inference finished")
      
      outputs.append((img_E, f'{filename}'))
      before_afters.append((input_img, img_E))
      
    return outputs, before_afters
    
def select_image(event: gr.SelectData, before_afters):
  index = event.index
  if index is None or index >= len(before_afters):
      return None
  return before_afters[index], index
  
def select_changed_image(index, before_afters):
  if index is None or index >= len(before_afters):
      return None
  return before_afters[index]
  
with gr.Blocks() as demo:
    gr.Markdown("# JPEG Artifacts Removal [FBCNN]")
    
    with gr.Row():
        input_img = gr.Gallery(
            label="Input Image(s)",
            file_types=['image'],
            type="filepath",
            height="auto"
        )

        output_img = gr.Gallery(
            label="Results",
            height="auto",
            interactive=False
        )
    
    is_gray = gr.Checkbox(label="Grayscale (Check this if your image is grayscale)")
    max_res = gr.Slider(1, 100, step=0.5, value=100, label="Output image resolution Percentage (Higher% = longer processing time)")
    input_quality = gr.Slider(1, 100, step=1, value=40, label="Intensity (Higher = stronger JPEG artifact removal)")
    
    run = gr.Button("Run")

    with gr.Row():
        before_afters = gr.State([])
        current_index = gr.State(0)
        before_after = gr.ImageSlider(label="Before/After (Select One)", value=None, height="auto")
        
    run.click(
        inference, 
        inputs=[input_img, is_gray, max_res, input_quality], 
        outputs=[output_img, before_afters]
    )
    
    output_img.select(
        select_image,
        inputs=[before_afters],
        outputs=[before_after, current_index]
    )
    
    output_img.change(
        select_changed_image,
        inputs=[current_index, before_afters],
        outputs=[before_after]
    )
    
    gr.Examples(
      examples=[
        [[("doraemon.jpg", "doraemon.jpg")], False, 100, 60],
        [[("tomandjerry.jpg", "tomandjerry.jpg")], False,  100, 60],
        [[("somepanda.jpg", "somepanda.jpg")], True,  100, 100],
        [[("cemetry.jpg", "cemetry.jpg")], False,  100, 70],
        [[("michelangelo_david.jpg", "michelangelo_david.jpg")], True,  100, 30],
        [[("elon_musk.jpg", "elon_musk.jpg")], False, 100, 45],
        [[("text.jpg", "text.jpg")], True,  100, 70]
      ], 
      inputs=[input_img, is_gray, max_res, input_quality], 
      outputs=[output_img, before_afters]
    )
    
    gr.Markdown("""
    JPEG Artifacts are noticeable distortions of images caused by JPEG lossy compression.
    Note that this is not an AI Upscaler, but just a JPEG Compression Artifact Remover.
    [Original Demo](https://huggingface.co/spaces/danielsapit/JPEG_Artifacts_Removal)
    [FBCNN GitHub Repo](https://github.com/jiaxi-jiang/FBCNN)  
    [Towards Flexible Blind JPEG Artifacts Removal (FBCNN, ICCV 2021)](https://arxiv.org/abs/2109.14573)  
    [Jiaxi Jiang](https://jiaxi-jiang.github.io/),  
    [Kai Zhang](https://cszn.github.io/),  
    [Radu Timofte](http://people.ee.ethz.ch/~timofter/)
    """)

demo.launch()