app.py

import gradio as gr
import sahi
import torch
from ultralyticsplus import YOLO, render_model_output

# Download images for the demo
sahi.utils.file.download_from_url(
    "https://raw.githubusercontent.com/kadirnar/dethub/main/data/images/highway.jpg",
    "highway.jpg",
)
sahi.utils.file.download_from_url(
    "https://raw.githubusercontent.com/obss/sahi/main/tests/data/small-vehicles1.jpeg",
    "small-vehicles1.jpeg",
)
sahi.utils.file.download_from_url(
    "https://raw.githubusercontent.com/ultralytics/yolov5/master/data/images/zidane.jpg",
    "zidane.jpg",
)

# Define available YOLOv8 segmentation models
model_names = [
    "yolov8n-seg.pt",
    "yolov8s-seg.pt",
    "yolov8m-seg.pt",
    "yolov8l-seg.pt",
    "yolov8x-seg.pt",
]

# Load the initial YOLOv8 model
current_model_name = "yolov8m-seg.pt"
model = YOLO(current_model_name)

def yolov8_inference(
    image: gr.Image = None,
    model_name: gr.Dropdown = None,
    image_size: gr.Slider = 640,
    conf_threshold: gr.Slider = 0.25,
    iou_threshold: gr.Slider = 0.45,
):
    """
    YOLOv8 inference function
    Args:
        image: Input image
        model_name: Name of the model
        image_size: Image size
        conf_threshold: Confidence threshold
        iou_threshold: IOU threshold
    Returns:
        Rendered image and mask coordinates with labels
    """
    global model
    global current_model_name
    # Switch model if a different one is selected
    if model_name != current_model_name:
        model = YOLO(model_name)
        current_model_name = model_name
    
    # Set model confidence and IOU thresholds
    model.overrides["conf"] = conf_threshold
    model.overrides["iou"] = iou_threshold
    
    # Perform inference with the YOLO model
    results = model.predict(image, imgsz=image_size, return_outputs=True)
    
    masks = []
    for result in results:
        masks.append([result.masks, result.labels])
    
    renders = []
    for image_results in results:
        render = render_model_output(
            model=model, image=image, model_output=image_results
        )
        renders.append(render)

    # Return mask coordinates and labels
    return masks

# Gradio app inputs and outputs
inputs = [
    gr.Image(type="filepath", label="Input Image"),
    gr.Dropdown(model_names, value=current_model_name, label="Model type"),
    gr.Slider(minimum=320, maximum=1280, value=640, step=32, label="Image Size"),
    gr.Slider(minimum=0.0, maximum=1.0, value=0.25, step=0.05, label="Confidence Threshold"),
    gr.Slider(minimum=0.0, maximum=1.0, value=0.45, step=0.05, label="IOU Threshold"),
]

outputs = gr.Textbox(label="Mask Coordinates and Labels")

# Example inputs for the Gradio app
examples = [
    ["zidane.jpg", "yolov8m-seg.pt", 640, 0.6, 0.45],
    ["highway.jpg", "yolov8m-seg.pt", 640, 0.25, 0.45],
    ["small-vehicles1.jpeg", "yolov8m-seg.pt", 640, 0.25, 0.45],
]

# Create the Gradio app interface
demo_app = gr.Interface(
    fn=yolov8_inference,
    inputs=inputs,
    outputs=outputs,
    title="Ultralytics YOLOv8 Segmentation Demo",
    examples=examples,
    cache_examples=True,
)

# Launch the Gradio app
demo_app.launch(
    debug=True,         # Show detailed errors in case of issues
    server_name="0.0.0.0",  # Host on all IPs
    server_port=7860,       # Custom port for accessing the app
    share=True              # To make the app accessible from a URL
)