Update app.py

app.py

@@ -1,59 +1,48 @@
-import gradio as gr
 import torch
-from bsrgan import BSRGAN
+from PIL import Image
+from RealESRGAN import RealESRGAN
+import gradio as gr
+
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+model2 = RealESRGAN(device, scale=2)
+model2.load_weights('weights/RealESRGAN_x2.pth', download=True)
+model4 = RealESRGAN(device, scale=4)
+model4.load_weights('weights/RealESRGAN_x4.pth', download=True)
+model8 = RealESRGAN(device, scale=8)
+model8.load_weights('weights/RealESRGAN_x8.pth', download=True)
+
 
-# Images
-torch.hub.download_url_to_file('https://raw.githubusercontent.com/kadirnar/bsrgan-pip/main/data/images/butterfly.png', 'butterfly.jpg')
+def inference(image: Image, size: str) -> Image:
+ try:
+ if size == '2x':
+ result = model2.predict(image.convert('RGB'))
+ elif size == '4x':
+ result = model4.predict(image.convert('RGB'))
+ else:
+ result = model8.predict(image.convert('RGB'))
+ except Exception as e:
+ print(e)
+ raise gr.Error(str(e))
+ return result
 
-def bsrgan_inference(
- image: gr.inputs.Image = None,
- model_path: gr.inputs.Dropdown = 'kadirnar/bsrgan',
-):
- """
- BSRGAN inference function
- Args:
- image: Input image
- model_path: Path to the model
- Returns:
- Rendered image
- """
- device = 'cuda:0' if torch.cuda.is_available() else 'cpu'
- model = BSRGAN(model_path, device=device, hf_model=True)
- pred = model.predict(img_path=image)
- return pred
- 
 
-inputs = [
- gr.inputs.Image(type="filepath", label="Input Image"),
- gr.inputs.Dropdown(
- label="Model",
- choices=[
- "kadirnar/bsrgan",
- "kadirnar/BSRGANx2",
- "kadirnar/RRDB_PSNR_x4",
- "kadirnar/RRDB_ESRGAN_x4",
- "kadirnar/DF2K",
- "kadirnar/DPED",
- "kadirnar/DF2K_JPEG",
- ],
- default="kadirnar/bsrgan",
- ),
-]
+title = "Face Real ESRGAN: 2x 4x 8x"
+description = "This is an unofficial demo for Real-ESRGAN. Scales the resolution of a photo. This model shows better results on faces compared to the original version.<br>Telegram BOT: https://t.me/restoration_photo_bot"
+article = "<div style='text-align: center;'>Twitter <a href='https://twitter.com/DoEvent' target='_blank'>Max Skobeev</a> | <a href='https://huggingface.co/sberbank-ai/Real-ESRGAN' target='_blank'>Model card</a> <center><img src='https://visitor-badge.glitch.me/badge?page_id=max_skobeev_face_esrgan' alt='visitor badge'></center></div>"
 
-outputs = gr.outputs.Image(type="filepath", label="Output Image")
-title = "BSRGAN: Designing a Practical Degradation Model for Deep Blind Image Super-Resolution."
-description = "BSRGAN for Deep Blind Image Super-Resolution model aims to design a practical degradation model for deep blind image super-resolution by considering the deterioration of image quality over time. It uses deep learning methods to predict the deterioration of image quality and to assist in the re-creation of images at higher resolution using these predictions."
-examples = [["butterfly.jpg", "kadirnar/bsrgan"]]
 
-demo_app = gr.Interface(
- fn=bsrgan_inference,
- inputs=inputs,
- outputs=outputs,
+gr.Interface(inference,
+ [gr.Image(type="pil"), 
+ gr.Radio(['2x', '4x', '8x'], 
+ type="value",
+ value='2x',
+ label='Resolution model')], 
+ gr.Image(type="pil", label="Output"),
  title=title,
  description=description,
- examples=examples,
- cache_examples=True,
- live=True,
- theme='huggingface',
-)
-demo_app.launch(debug=True, enable_queue=True)
+ article=article,
+ examples=[['groot.jpeg', "2x"]],
+ allow_flagging='never',
+ theme="default",
+ cache_examples=False,
+ ).queue().launch(show_error=True)