import gradio as gr
from gradio_imageslider import ImageSlider
from PIL import Image
import numpy as np
from aura_sr import AuraSR
import torch
import os
import time
import platform
import argparse
# Global variable to control batch processing cancellation.
stop_batch_flag = False
def open_folder():
open_folder_path = os.path.abspath("outputs")
if platform.system() == "Windows":
os.startfile(open_folder_path)
elif platform.system() == "Linux":
os.system(f'xdg-open "{open_folder_path}"')
def get_placeholder_image():
"""
Creates a placeholder image (if not already present) and returns its file path.
This placeholder is a blank (white) image that will be used for progress updates.
"""
placeholder_path = "placeholder.png"
if not os.path.exists(placeholder_path):
placeholder = Image.new("RGB", (256, 256), (255, 255, 255))
placeholder.save(placeholder_path)
return placeholder_path
# Force CPU usage
torch.set_default_tensor_type(torch.FloatTensor)
# Override torch.load to always use CPU
original_load = torch.load
torch.load = lambda *args, **kwargs: original_load(*args, **kwargs, map_location=torch.device('cpu'))
# Initialize the AuraSR model
aura_sr = AuraSR.from_pretrained("fal/AuraSR-v2")
# Restore original torch.load
torch.load = original_load
def process_single_image(input_image_path, reduce_seams):
if input_image_path is None:
raise gr.Error("Please provide an image to upscale.")
# Send an initial progress update.
# Instead of (None, None), we use the placeholder image file paths.
placeholder = get_placeholder_image()
yield [(placeholder, placeholder), "Starting upscaling..."]
# Load the image.
pil_image = Image.open(input_image_path)
# Upscale using the chosen method.
start_time = time.time()
if reduce_seams:
print("using reduce seams")
upscaled_image = aura_sr.upscale_4x_overlapped(pil_image)
else:
upscaled_image = aura_sr.upscale_4x(pil_image)
processing_time = time.time() - start_time
print(f"Processing time: {processing_time:.2f} seconds")
# Save the upscaled image.
output_folder = "outputs"
os.makedirs(output_folder, exist_ok=True)
input_filename = os.path.basename(input_image_path)
output_filename = os.path.splitext(input_filename)[0]
output_path = os.path.join(output_folder, output_filename + ".png")
counter = 1
while os.path.exists(output_path):
output_path = os.path.join(output_folder, f"{output_filename}_{counter:04d}.png")
counter += 1
upscaled_image.save(output_path)
# Send the final progress update along with the before/after slider images.
yield [(input_image_path, output_path),
f"Upscaling complete in {processing_time:.2f} seconds"]
def process_batch(input_folder, output_folder=None, reduce_seams=False):
global stop_batch_flag
# Reset the stop flag for each new batch process.
stop_batch_flag = False
if not input_folder:
raise gr.Error("Please provide an input folder path.")
if not output_folder:
output_folder = "outputs"
os.makedirs(output_folder, exist_ok=True)
input_files = [f for f in os.listdir(input_folder) if f.lower().endswith(
('.png', '.jpg', '.jpeg', '.bmp', '.tiff'))]
total_files = len(input_files)
processed_files = 0
results = []
yield [results, "Starting batch processing..."]
for filename in input_files:
# Check if the stop flag has been set.
if stop_batch_flag:
yield [results, "Batch processing cancelled by user."]
return
input_path = os.path.join(input_folder, filename)
pil_image = Image.open(input_path)
start_time = time.time()
if reduce_seams:
upscaled_image = aura_sr.upscale_4x_overlapped(pil_image)
else:
upscaled_image = aura_sr.upscale_4x(pil_image)
processing_time = time.time() - start_time
output_filename = os.path.splitext(filename)[0] + ".png"
output_path = os.path.join(output_folder, output_filename)
counter = 1
while os.path.exists(output_path):
output_path = os.path.join(output_folder, f"{os.path.splitext(filename)[0]}_{counter:04d}.png")
counter += 1
upscaled_image.save(output_path)
processed_files += 1
results.append(output_path)
yield [results, f"Processed {processed_files}/{total_files}: {filename} in {processing_time:.2f} seconds"]
yield [results, f"Batch processing complete. {processed_files} images processed."]
def stop_batch_process():
global stop_batch_flag
stop_batch_flag = True
return "Stop button clicked. Cancelling batch processing..."
title = """AuraSR Giga Upscaler V4 by SECourses - Upscales to 4x
AuraSR: new open source super-resolution upscaler based on GigaGAN. Works perfect on some images and fails on some images so give it a try
Works very fast and very VRAM friendly
"""
def create_demo():
with gr.Blocks() as demo:
gr.HTML(title)
with gr.Tab("Single Image"):
with gr.Row():
with gr.Column(scale=1):
input_image = gr.Image(label="Input Image", type="filepath")
reduce_seams = gr.Checkbox(
label="Reduce Seam Artifacts",
value=True,
info="upscale_4x upscales the image in tiles that do not overlap. This can result in seams. Use upscale_4x_overlapped to reduce seams. This will double the time upscaling by taking an additional pass and averaging the results."
)
process_btn = gr.Button(value="Upscale Image", variant="primary")
with gr.Column(scale=1):
# Use the ImageSlider component for comparing before & after images.
# "filepath" type means the component expects image file paths.
output_slider = ImageSlider(label="Before / After", type="filepath", slider_color="blue")
progress_text = gr.Markdown("Progress messages will appear here.")
btn_open_outputs = gr.Button("Open Outputs Folder", variant="primary")
btn_open_outputs.click(fn=open_folder)
process_btn.click(
fn=process_single_image,
inputs=[input_image, reduce_seams],
outputs=[output_slider, progress_text]
)
with gr.Tab("Batch Processing"):
with gr.Row():
input_folder = gr.Textbox(label="Input Folder Path")
output_folder = gr.Textbox(label="Output Folder Path (Optional)")
reduce_seams_batch = gr.Checkbox(
label="Reduce Seam Artifacts",
value=True,
info="upscale_4x upscales the image in tiles that do not overlap. This can result in seams. Use upscale_4x_overlapped to reduce seams. This will double the time upscaling by taking an additional pass and averaging the results."
)
with gr.Row():
batch_process_btn = gr.Button(value="Process Batch", variant="primary")
stop_batch_btn = gr.Button(value="Stop Batch Processing", variant="secondary")
with gr.Column():
output_gallery_batch = gr.Gallery(label="Processed Images")
progress_text_batch = gr.Markdown("Progress messages will appear here.")
batch_process_btn.click(
fn=process_batch,
inputs=[input_folder, output_folder, reduce_seams_batch],
outputs=[output_gallery_batch, progress_text_batch]
)
stop_batch_btn.click(
fn=stop_batch_process,
inputs=[],
outputs=[progress_text_batch]
)
return demo
if __name__ == "__main__":
parser = argparse.ArgumentParser(description="AuraSR Image Upscaling")
parser.add_argument("--share", action="store_true", help="Create a publicly shareable link")
args = parser.parse_args()
demo = create_demo()
demo.launch(debug=True, inbrowser=True, share=args.share)