app.py

import numpy as np
import gradio as gr
from ultralytics import YOLO
import tempfile
import cv2

def inference(image, video, model_id, image_size, conf_threshold):
    if model_id == "yolov10n-obb":
        model = YOLO("")
    elif model_id == "yolov10s-obb":
        model = YOLO("")
    elif model_id == "yolov10m-obb":
        model = YOLO("")
    
    if image:
        results = model.predict(source=image, imgsz=image_size, conf=conf_threshold, device="cpu")
        annotated_image = results[0].plot()
        return annotated_image[:, :, ::-1], None
    else:
        video_path = tempfile.mktemp(suffix=".webm")
        with open(video_path, "wb") as f:
            with open(video, "rb") as g:
                f.write(g.read())

        cap = cv2.VideoCapture(video_path)
        fps = cap.get(cv2.CAP_PROP_FPS)
        frame_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
        frame_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))

        output_video_path = tempfile.mktemp(suffix=".webm")
        out = cv2.VideoWriter(output_video_path, cv2.VideoWriter_fourcc(*'vp90'), fps, (frame_width, frame_height))

        while cap.isOpened():
            ret, frame = cap.read()
            if not ret:
                break

            results = model.predict(source=frame, imgsz=image_size, conf=conf_threshold)
            annotated_frame = results[0].plot()
            out.write(annotated_frame)

        cap.release()
        out.release()

        return None, output_video_path

def inference_for_examples(image, model_path, image_size, conf_threshold):
    annotated_image, _ = inference(image, None, model_path, image_size, conf_threshold)
    return annotated_image

def app():
    with gr.Blocks():
        with gr.Row():
            with gr.Column():
                image = gr.Image(type="pil", label="Image", visible=True)
                video = gr.Video(label="Video", visible=False)
                input_type = gr.Radio(
                    choices=["Image", "Video"],
                    value="Image",
                    label="Input Type",
                )
                model_id = gr.Dropdown(
                    label="Model",
                    choices=[
                        "yolov10n-obb",
                        "yolov10s-obb",
                        "yolov10m-obb",
                    ],
                    value="yolov10n-obb",
                )
                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,
                )
                inferBtn = gr.Button(value="Detect")

            with gr.Column():
                output_image = gr.Image(type="numpy", label="Annotated Image", visible=True)
                output_video = gr.Video(label="Annotated Video", visible=False)

        def update_visibility(input_type):
            image = gr.update(visible=True) if input_type == "Image" else gr.update(visible=False)
            video = gr.update(visible=False) if input_type == "Image" else gr.update(visible=True)
            output_image = gr.update(visible=True) if input_type == "Image" else gr.update(visible=False)
            output_video = gr.update(visible=False) if input_type == "Image" else gr.update(visible=True)

            return image, video, output_image, output_video

        input_type.change(
            fn=update_visibility,
            inputs=[input_type],
            outputs=[image, video, output_image, output_video],
        )

        def run_inference(image, video, model_id, image_size, conf_threshold, input_type):
            if input_type == "Image":
                return inference(image, None, model_id, image_size, conf_threshold)
            else:
                return inference(None, video, model_id, image_size, conf_threshold)


        inferBtn.click(
            fn=run_inference,
            inputs=[image, video, model_id, image_size, conf_threshold, input_type],
            outputs=[output_image, output_video],
        )

        gr.Examples(
            examples=[
                [
                    "test_images/P0024.jpg",
                    "yolov10n-obb",
                    1024,
                    0.25,
                ],
                [
                    "test_images/P0035.jpg",
                    "yolov10n-obb",
                    1024,
                    0.25,
                ],
                [
                    "test_images/P0121.jpg",
                    "yolov10n-obb",
                    1024,
                    0.25,
                ],
                [
                    "test_images/P0180.jpg",
                    "yolov10n-obb",
                    1024,
                    0.25,
                ],
                [
                    "test_images/P0279.jpg",
                    "yolov10n-obb",
                    1024,
                    0.25,
                ],
                [
                    "test_images/P2112.jpg",
                    "yolov10n-obb",
                    1024,
                    0.25,
                ],               
            ],
            fn=inference_for_examples,
            inputs=[
                image,
                model_id,
                image_size,
                conf_threshold,
            ],
            outputs=[output_image],
            cache_examples='lazy',
        )

gradio_app = gr.Blocks()
with gradio_app:
    gr.Markdown(
        """
    # YOLOv10 - OBB (Oriented Bounding Box)

    for more detail description about this model, please visit [here](https://github.com/hamhanry/YOLOv10-OBB)
        """
    )
    with gr.Row():
        with gr.Column():
            app()
if __name__ == '__main__':
    gradio_app.queue()
    gradio_app.launch()