README.md

sd-webui-rembg

Fork from stable-diffusion-webui-rembg, But add API Support.

Installation

Choose one of the following methods, Need to use webui with extension support ^{(Versions after 2023)}

Method 1

Use the Install from URL provided by webui to install

Click in order Extensions - Install from URL

Then fill in the first text box with https://github.com/hicoldcat/sd-webui-rembg.git, click the Install button.

After that, switch to the Installed panel and click the Apply and restart UI button.

Method 2

Clone to your extension directory manually.

git clone https://github.com/hicoldcat/sd-webui-rembg.git extensions/sd-webui-rembg

API Support

API Usage

We have added an API endpoint to allow for automated workflows.

The API utilizes both Segment Anything and GroundingDINO to return masks of all instances of whatever object is specified in the text prompt.

This is an extension of the existing Stable Diffusion Web UI API.

There are 2 endpoints exposed

• GET sam-webui/heartbeat
• POST /sam-webui/image-mask

The heartbeat endpoint can be used to ensure that the API is up.

The image-mask endpoint accepts a payload that includes your base64-encoded image.

Below is an example of how to interface with the API using requests.

API Example

import base64
import requests
from PIL import Image
from io import BytesIO
url = "http://127.0.0.1:7860/sam-webui/image-mask"
def image_to_base64(img_path: str) -> str:
    with open(img_path, "rb") as img_file:
        img_base64 = base64.b64encode(img_file.read()).decode()
    return img_base64
payload = {
    "image": image_to_base64("IMAGE_FILE_PATH"),
    "prompt": "TEXT PROMPT",
    "box_threshold": 0.3,
    "padding": 30 #Optional param to pad masks
}
res = requests.post(url, json=payload)
for dct in res.json():
    image_data = base64.b64decode(dct['image'])
    image = Image.open(BytesIO(image_data))
    image.show()

Rembg

Extension for webui. Removes backgrounds from pictures.

Find the UI for rembg in the Extras tab after installing the extension.

Credits

• rembg library that does all the work: https://github.com/danielgatis/rembg