Face2Face

Face2Face is a generative AI technology to swap faces (aka Deep Fake) in images from one to another. For example you can swap your face with Mona Lisa our your favourite superstar. With this repository you can:

Swap faces from one image to another. Powered by Insightface
Create face embeddings. With these embeddings you can later swap faces without running the whole stack again.

All of this is wrapped into a convenient REST API with FAST_API

Disclaimer

I'm not responsible of any misuse of the repository. Face swapping is a powerful technology that can be used for good and bad. Please use it responsibly and do not harm others. Do not publish any images without the consent of the people in the images. The credits for face swapping technology go to the great Insightface Team thank you insightface.ai. This projects uses their pretrained models and code. Special thanks goes to their work around ROOP. Some parts of the code stem from the ROOP repository.

I do not claim any authorship for this repository. My contribution was simply wrapping the code into a REST Api.

Setup

Clone the repository.
(Optional) Create a virtual environment. With python -m venv venv and activate it with venv/Scripts/activate.
Install the requirements. pip install -r requirements.txt
Don't forget to install pytorch gpu version (with pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118)

Usage

Start the server by running the provided .bat file "start_server.bat" 2. or by using python face_swapper_REST/server.py --port 8020 make sure the python PYTHONPATH is set to the root of this repository.
To test the server, open http://localhost:8020/docs in your browser.

Then make post requests to the server with your favorite tool or library. Here are some examples to inference with a python client.

Note: The first time you start the server, it will download the models. This can take a while. If this fails, you can download the files manually and store them in models/ or models/insightface/inswapper_128.onnx

For face swapping

import requests

# load images from disc
with open("src.jpg", "rb") as image:
    src_img = image.read()
with open("target.jpg", "rb") as image:
    target_img = image.read()

# send post request
response = requests.post("http://localhost:8020/swap_one", files={"source_img": src_img, "target_img": target_img})

The response is a .png file as bytes. Use the following code to convert back.

import cv2
from io import BytesIO

# convert to image file
swapped = cv2.imread(BytesIO(response.content))

For face embedding generation

import requests

with open("src.jpg", "rb") as image:
    src_img = image.read()

response = requests.post("http://localhost:8020/add_reference_face", params={ "face_name": "myface", "save": True}, files={"source_img": src_img})

The response is a .npz file as bytes. After the embedding was created it can be used in the next swapping with the given face_name.

For face swapping with saved reference faces

import requests
with open("target.jpg", "rb") as image:
    target_img = image.read()
response = requests.post("http://localhost:8020/swap_from_reference_face", params={ "face_name" : "myface"}, files={"target_img": target_img})

In this example it is assumed that previously a face embedding with name "myface" was created with the add_reference_face endpoint.

Inference via script

You can also use the class FaceSwapper to do the inference.

from face_swapper_REST.face_swapper import FaceSwapper
swapper = FaceSwapper()
swapped = swapper.swap_one("src.jpg", "target.jpg")

Contribute

Any help with maintaining and extending the package is welcome. Feel free to open an issue or a pull request. ToDo: make inference faster by implementing batching.

sauravgarg540/face2face