app.py

import gradio as gr
import cv2
from huggingface_hub import hf_hub_download
import yolov9

def detect_objects_on_video(video_path, model_path, interval, image_size, conf_threshold, iou_threshold):
    cap = cv2.VideoCapture(video_path)
    count = 0
    detections = []
    while cap.isOpened():
        ret, frame = cap.read()
        if not ret:
            break

        count += 1
        if count % interval == 0:
            # Perform object detection on the frame
            model = yolov9.load(model_path)
            model.conf = conf_threshold
            model.iou = iou_threshold
            results = model(frame, size=image_size)
            
            # Optionally, show detection bounding boxes on image
            output = results.render()
            detections.append(output[0])

    cap.release()
    cv2.destroyAllWindows()
    return detections

def app():
    with gr.Blocks():
        with gr.Row():
            with gr.Column():
                video_path = gr.Video(type="file", label="Video")
                model_path = gr.Dropdown(
                    label="Model",
                    choices=[
                        "best.pt",
                    ],
                    value="./best.pt",
                )
                interval = gr.Number(label="Screenshot Interval (seconds)", default=30, step=1)
                image_size = gr.Slider(
                    label="Image Size",
                    minimum=320,
                    maximum=1280,
                    step=32,
                    value=640,
                )
                conf_threshold = gr.Slider(
                    label="Confidence Threshold",
                    minimum=0.1,
                    maximum=1.0,
                    step=0.1,
                    value=0.4,
                )
                iou_threshold = gr.Slider(
                    label="IoU Threshold",
                    minimum=0.1,
                    maximum=1.0,
                    step=0.1,
                    value=0.5,
                )
                yolov9_infer = gr.Button(value="Detect Objects")

            with gr.Column():
                output_images = gr.Image(type="numpy", label="Output Images")

        yolov9_infer.click(
            fn=detect_objects_on_video,
            inputs=[
                video_path,
                model_path,
                interval,
                image_size,
                conf_threshold,
                iou_threshold,
            ],
            outputs=[output_images],
        )

gradio_app = gr.Blocks()
with gradio_app:
    gr.HTML(
        """
    <h1 style='text-align: center'>
    YOLOv9: Detect Objects in Video
    </h1>
    """
    )
    with gr.Row():
        with gr.Column():
            app()

gradio_app.launch(debug=True)