app.py

import gradio as gr
from ultralytics import YOLO
import spaces
import torch

# Load pre-trained YOLOv8 model
model = YOLO("yolov8x-doclaynet-epoch64-imgsz640-initiallr1e-4-finallr1e-5.pt")

# Get class names from model
class_names = model.names

@spaces.GPU(duration=60)
def process_image(image):
    try:
        # Process the image
        results = model(source=image, save=False, show_labels=True, show_conf=True, show_boxes=True)
        result = results[0]  # Get the first result

        # Extract annotated image and labels with class names
        annotated_image = result.plot()

        # Convert box.cls tensor to float before formatting
        detected_areas_labels = "\n".join([
            f"{class_names[int(box.cls.item())].upper()}: {box.conf:.2f}" for box in result.boxes
        ])

        return annotated_image, detected_areas_labels
    except Exception as e:
        return None, f"Error processing image: {e}"


# Create the Gradio Interface
with gr.Blocks() as demo:
    gr.Markdown("# Document Segmentation Demo (ZeroGPU)")  
    # Input Components
    input_image = gr.Image(type="pil", label="Upload Image")

    # Output Components
    output_image = gr.Image(type="pil", label="Annotated Image")
    output_text = gr.Textbox(label="Detected Areas and Labels")

    # Button to trigger inference
    btn = gr.Button("Run Document Segmentation")
    btn.click(fn=process_image, inputs=input_image, outputs=[output_image, output_text])

# Launch the demo
demo.queue(max_size=1).launch()  # Queue to handle concurrent requests