app.py

import gradio as gr
import cv2

from ultralyticsplus import YOLO, render_result

# Model Heading and Description
model_heading = "ProductDetect-o-Matic: Shelf Wizard for Retail Magic"
description = """ 🛒 Prepare to be amazed by ProductDetect-o-Matic! 🪄 Unveil the secrets of your shelves and keep products in check. From the mystical 'Empty Shelf' to the enigmatic 'Magical Product', our model is here to bring retail magic to life. Created by Foduu AI, your ultimate sidekick for retail success. 💼🎩✨
🛍️ Ready to experience the retail revolution? Reach out at [email protected] and let's make your shelves enchanting! Giving a thumbs up won't make you a retail wizard, but it's a step closer to making your shelves smile! 🚀👍🪄
📧 Contact us: [email protected]
👍 Like | Join the Retail Adventure!"""

image_path = [
    ['test/test1.jpg', 'foduucom/product-detection-in-shelf-yolov8', 640, 0.25, 0.45],
    ['test/test2.jpg', 'foduucom/product-detection-in-shelf-yolov8', 640, 0.25, 0.45]
]

# Load YOLO model
model = YOLO('foduucom/product-detection-in-shelf-yolov8')

############################################################# Image Inference ############################################################
def yolov8_img_inference(
    image: gr.inputs.Image = None,
    model_path: gr.inputs.Dropdown = None,
    image_size: gr.inputs.Slider = 640,
    conf_threshold: gr.inputs.Slider = 0.25,
    iou_threshold: gr.inputs.Slider = 0.45,
):
    model = YOLO(model_path)
    model.overrides['conf'] = conf_threshold
    model.overrides['iou'] = iou_threshold
    model.overrides['agnostic_nms'] = False
    model.overrides['max_det'] = 1000
    results = model.predict(image)
    render = render_result(model=model, image=image, result=results[0])

    return render


inputs_image = [
    gr.inputs.Image(type="filepath", label="Input Image"),
    gr.inputs.Dropdown(["foduucom/product-detection-in-shelf-yolov8"], default="foduucom/product-detection-in-shelf-yolov8", label="Model"),
    gr.inputs.Slider(minimum=320, maximum=1280, default=640, step=32, label="Image Size"),
    gr.inputs.Slider(minimum=0.0, maximum=1.0, default=0.25, step=0.05, label="Confidence Threshold"),
    gr.inputs.Slider(minimum=0.0, maximum=1.0, default=0.45, step=0.05, label="IOU Threshold"),
]

outputs_image = gr.outputs.Image(type="filepath", label="Output Image")
interface_image = gr.Interface(
    fn=yolov8_img_inference,
    inputs=inputs_image,
    outputs=outputs_image,
    title=model_heading,
    description=description,
    examples=image_path,
    cache_examples=False,
    theme='huggingface'
)

################################################## Video Inference ############################################################
def show_preds_video(
    video_path: str = None,
    model_path: str = None,
    image_size: int = 640,
    conf_threshold: float = 0.25,
    iou_threshold: float = 0.45,
):
    cap = cv2.VideoCapture(video_path)

    while cap.isOpened():
        success, frame = cap.read()

        if success:
            model = YOLO(model_path)
            model.overrides['conf'] = conf_threshold
            model.overrides['iou'] = iou_threshold
            model.overrides['agnostic_nms'] = False
            model.overrides['max_det'] = 1000
            results = model.predict(frame)
            annotated_frame = results[0].plot()

            if cv2.waitKey(1) & 0xFF == ord("q"):
                break
        else:
            break

    cap.release()
    cv2.destroyAllWindows()


inputs_video = [
    gr.components.Video(type="filepath", label="Input Video"),
    gr.inputs.Dropdown(["foduucom/product-detection-in-shelf-yolov8"], default="foduucom/product-detection-in-shelf-yolov8", label="Model"),
    gr.inputs.Slider(minimum=320, maximum=1280, default=640, step=32, label="Image Size"),
    gr.inputs.Slider(minimum=0.0, maximum=1.0, default=0.25, step=0.05, label="Confidence Threshold"),
    gr.inputs.Slider(minimum=0.0, maximum=1.0, default=0.45, step=0.05, label="IOU Threshold"),
]

outputs_video = gr.outputs.Image(type="filepath", label="Output Video")
video_path = [['test/testvideo.mp4', 'foduucom/product-detection-in-shelf-yolov8', 640, 0.25, 0.45]]
interface_video = gr.Interface(
    fn=show_preds_video,
    inputs=inputs_video,
    outputs=outputs_video,
    title=model_heading,
    description=description,
    examples=video_path,
    cache_examples=False,
    theme='huggingface'
)

gr.TabbedInterface(
    [interface_image, interface_video],
    tab_names=['Image inference', 'Video inference']
).queue().launch()