app.py

import PIL.Image
import gradio as gr
from PIL import ImageColor
from rembg import new_session, remove

from utils import *

remove_bg_models = {
    "U2NET": "u2net",
    "U2NET Human Seg": "u2net_human_seg",
    "U2NET Cloth Seg": "u2net_cloth_seg"
}

model_choices = keys(remove_bg_models)


def alpha_matting(state):
    if state:
        return 270, 20, 11
    return 0, 0, 0


def predict(image,
            session,
            matting,
            only_mask,
            post_process_mask,
            foreground_threshold,
            background_threshold,
            matting_erode_size,
            new_bg_color,
            bg_color,
            transparency):
    session = new_session(remove_bg_models[session])

    if new_bg_color:
        r, g, b, _ = ImageColor.getcolor(bg_color, "RGBA")
        bg_color = r, g, b, transparency
    else:
        bg_color = None

    return remove(data=image,
                  session=session,
                  alpha_matting=matting,
                  only_mask=only_mask,
                  post_process_mask=post_process_mask,
                  bgcolor=bg_color,
                  alpha_matting_foreground_threshold=foreground_threshold,
                  alpha_matting_background_threshold=background_threshold,
                  alpha_matting_erode_size=matting_erode_size)


footer = r"""
<center>
<b>
Demo based on <a href='https://github.com/danielgatis/rembg'>Rembg</a>
</b>
</center>
"""

with gr.Blocks(title="Remove background") as app:
    gr.HTML("<center><h1>Remove Background Tool</h1></center>")
    with gr.Row(equal_height=False):
        with gr.Column():
            input_img = gr.Image(type="numpy", label="Input image")
            drp_models = gr.Dropdown(choices=model_choices, label="Model Segment", value="U2NET")
            with gr.Row():
                chk_alm = gr.Checkbox(label="Alpha Matting", value=False)
                chk_psm = gr.Checkbox(label="Post process mask", value=False)
                chk_msk = gr.Checkbox(label="Only Mask", value=False)
            sld_aft = gr.Slider(0, 300, value=0, step=1, label="Alpha matting foreground threshold")
            sld_amb = gr.Slider(0, 50, value=0, step=1, label="Alpha matting background threshold")
            sld_aes = gr.Slider(0, 20, value=0, step=1, label="Alpha matting erode size")
            with gr.Box():
                with gr.Row():
                    chk_col = gr.Checkbox(label="Change background color", value=False)
                    color = gr.ColorPicker(label="Pick a new color")
                    trans = gr.Number(label="Transparency level", value=255, precision=0, minimum=0, maximum=255)
            run_btn = gr.Button(value="Remove background", variant="primary")
        with gr.Column():
            output_img = gr.Image(type="pil", label="result")
            gr.ClearButton(components=[input_img, output_img])

    chk_alm.change(alpha_matting, inputs=[chk_alm], outputs=[sld_aft, sld_amb, sld_aes])

    run_btn.click(predict, [input_img, drp_models,
                            chk_alm, chk_msk, chk_psm,
                            sld_aft, sld_amb, sld_aes, chk_col, color, trans], [output_img])

    with gr.Row():
        gr.HTML(footer)

app.launch(share=False, debug=True, enable_queue=True, show_error=True)