app.py

import gradio as gr
from PIL import Image, ImageDraw
import yolov5
import json

model = yolov5.load("./best.pt")

def yolo(im):

    results = model(im)  # inference

    df = results.pandas().xyxy[0].to_json(orient="records")

    res = json.loads(df)

    draw = ImageDraw.Draw(im)

    for bb in res:
        xmin = bb['xmin']
        ymin = bb['ymin']
        xmax = bb['xmax']
        ymax = bb['ymax']
        draw.rectangle([xmin, ymin, xmax, ymax], outline="red", width=3)

    return [
        res,
        im,
    ]


inputs = [
    gr.Image(type='pil', label="Original Image")
]
outputs = [
    gr.JSON(label="Output JSON"),
    gr.Image(type='pil', label="Output Image with Boxes"),
]

title = "YOLOv5 Face"
description = """<p>YOLOv5 Face Gradio demo for object detection. Upload an image or click an example image to use.</p>"""

article = """
<ul>
<li>『文正草子』（日本古典籍データセット（国文研所蔵）CODH配信）</li>
<li>『源氏百人一首』(大阪公立大学中百舌鳥図書館 国文学研究資料館)</li>
<li>『源氏物語』(国文学研究資料館)</li>
<li>『百鬼夜行図』（東京大学）</li>
</ul>
<p style='text-align: center'>YOLOv5 Face is an object detection model trained on the <a href=\"https://doi.org/10.20676/00000353\">顔コレデータセット</a>.</p>"""

examples = [
    ["『文正草子』（日本古典籍データセット（国文研所蔵）CODH配信）.jpg"],
    ['『源氏百人一首』(大阪公立大学中百舌鳥図書館 国文学研究資料館).jpg'],
    ['『源氏物語』(国文学研究資料館).jpg'],
    ['『百鬼夜行図』（東京大学）.jpg']
]
demo = gr.Interface(
    fn=yolo, 
    inputs=inputs,
    outputs=outputs, 
    title=title, 
    description=description, 
    article=article, 
    examples=examples
    )

demo.css = """
.json-holder {
    height: 300px;
    overflow: auto;
}
"""

demo.launch(share=False)