app.py

import gradio as gr
import spaces
from PIL import Image
from ultralytics import YOLO

# Load Models
models = {
    "yolov10n": YOLO("jameslahm/yolov10n"),
    "yolov10s": YOLO("jameslahm/yolov10s"),
    "yolov10m": YOLO("jameslahm/yolov10m"),
    "yolov10b": YOLO("jameslahm/yolov10b"),
    "yolov10l": YOLO("jameslahm/yolov10l"),
    "yolov10x": YOLO("jameslahm/yolov10x"),
}

@spaces.GPU(duration=30)
def yolov10_inference(image, model_id, image_size, conf_threshold, iou_threshold):
    model = models[model_id]
    results = model.predict(
        source=image,
        imgsz=image_size,
        conf=conf_threshold,
        iou=iou_threshold,
    )
    annotated_image = results[0].plot()
    return Image.fromarray(annotated_image[..., ::-1])

def app():
    with gr.Blocks():
        with gr.Row():
            with gr.Column():
                image = gr.Image(type="pil", label="Image")
                model_id = gr.Dropdown(
                    label="Model",
                    choices=[
                        "yolov10n",
                        "yolov10s",
                        "yolov10m",
                        "yolov10b",
                        "yolov10l",
                        "yolov10x",
                    ],
                    value="yolov10m",
                )
                image_size = gr.Slider(
                    label="Image Size",
                    minimum=320,
                    maximum=1280,
                    step=32,
                    value=640,
                )
                conf_threshold = gr.Slider(
                    label="Confidence Threshold",
                    minimum=0.0,
                    maximum=1.0,
                    step=0.05,
                    value=0.25,
                )
                iou_threshold = gr.Slider(
                    label="IoU Threshold",
                    minimum=0.0,
                    maximum=1.0,
                    step=0.05,
                    value=0.45,
                )
                yolov10_infer = gr.Button(value="Detect Objects")

            with gr.Column():
                output_image = gr.Image(type="pil", label="Annotated Image")

    gr.Examples(
        examples=[
            ["dog.jpeg", "yolov10m", 640, 0.25, 0.45],
            ["huggingface.jpg", "yolov10m", 640, 0.25, 0.45],
            ["zidane.jpg", "yolov10m", 640, 0.25, 0.45],
        ],
        fn=yolov10_inference,
        inputs=[image, model_id, image_size, conf_threshold, iou_threshold],
        outputs=[output_image],
        cache_examples='lazy',
    )

    yolov10_infer.click(
        fn=yolov10_inference,
        inputs=[image, model_id, image_size, conf_threshold, iou_threshold],
        outputs=[output_image],
    )

gradio_app = gr.Blocks()

with gradio_app:
    with gr.Row():
        with gr.Column():
            gr.HTML(
                """
            <h1 style='text-align: center'>
            YOLOv10: Real-Time End-to-End Object Detection
            </h1>
            """
            )
            gr.HTML(
                """
                <h3 style='text-align: center'>
                <a href='https://arxiv.org/abs/2405.14458' target='_blank'>arXiv</a> | <a href='https://github.com/THU-MIG/yolov10' target='_blank'>github</a>
                </h3>
                """
            )
            gr.HTML(
                """
            <h3 style='text-align: center'>
            Follow me for more!
            </h3>
            """
            )
            gr.HTML(
                """
                <div style='text-align: center;'>
                Twitter | Github | Linkedin | HuggingFace
                </div>
                """
            )

    app()

if __name__ == "__main__":
    gradio_app.launch()