| --- |
| license: apache-2.0 |
| title: FACE SWAP API |
| sdk: docker |
| emoji: 🏃 |
| colorFrom: indigo |
| colorTo: yellow |
| short_description: SWAP FACES ONLINE VIA API |
| --- |
| # inswapper-flask-api |
|
|
| Python Flask API for Face Swapper and Restoration |
| powered by [insightface](https://github.com/deepinsight/insightface). |
|
|
| ## Installation |
|
|
| ### Clone this repository |
|
|
| ```bash |
| git clone https://github.com/ashleykleynhans/inswapper-flask-api.git |
| cd inswapper-flask-api |
| ``` |
|
|
| ### Install the required Python dependencies |
|
|
| #### Linux and Mac |
|
|
| ``` |
| python3 -m venv venv |
| source venv/bin/activate |
| pip3 install -r requirements.txt |
| ``` |
|
|
| #### Windows |
|
|
| ``` |
| python3 -m venv venv |
| venv\Scripts\activate |
| pip3 install -r requirements.txt |
| ``` |
|
|
| ## Download Checkpoints |
|
|
| You will need to download the [face swap model]( |
| https://huggingface.co/ashleykleynhans/inswapper/resolve/main/inswapper_128.onnx) and |
| save it under `./checkpoints`. |
| |
| ```bash |
| mkdir checkpoints |
| wget -O ./checkpoints/inswapper_128.onnx https://huggingface.co/ashleykleynhans/inswapper/resolve/main/inswapper_128.onnx |
| ``` |
| |
| ## Face Restoration |
| |
| In order to obtain better results, it is highly recommended to enable |
| a face restoration model, which will improve image quality. |
| This application uses [CodeFormer](https://github.com/sczhou/CodeFormer) |
| for face restoration. |
| |
| The required models will be downloaded automatically the first time |
| the face swap API is called. |
| |
| You will require [Git Large File Storage]( |
| https://docs.github.com/en/repositories/working-with-files/managing-large-files/installing-git-large-file-storage) |
| to be installed on your system before you can run the following command. |
| |
| ```bash |
| git lfs install |
| git clone https://huggingface.co/spaces/sczhou/CodeFormer |
| ``` |
| |
| ## Examples |
| |
| Refer to the [examples](./examples) provided for getting started |
| with making calls to the API. |
| |
| ## Benchmarks |
| |
| These benchmarks are for a source image with a resolution of 960x1280 |
| and a target image with a resolution of 1200x750, upscaled by 1 and |
| with CodeFormer Face Restoration enabled. |
| |
| | System | Time Taken | |
| |---------------------------------------------|---------------| |
| | macOS Ventura 13.4.1 on Apple M1 Max | 68.6 seconds | |
| | Ubuntu 22.04 LTS on t3a.xlarge AWS instance | 248.1 seconds | |
| | Ubuntu 22.04 LTS on an A5000 RunPod GPU pod | 14.2 seconds | |
| | Windows 10 | 103.9 seconds | |
| |
| Get a [RunPod](https://runpod.io?ref=w18gds2n) account. |
| |
| ## Acknowledgements |
| |
| 1. Thanks [insightface.ai](https://insightface.ai/) for releasing their powerful swap model that made this possible. |
| 2. This codebase is built on top of [inswapper](https://github.com/haofanwang/inswapper) and [CodeFormer]( |
| https://huggingface.co/spaces/sczhou/CodeFormer). |
| 3. [inswapper](https://github.com/haofanwang/inswapper) is built on the top of [sd-webui-roop]( |
| https://github.com/s0md3v/sd-webui-roop) and [CodeFormer](https://huggingface.co/spaces/sczhou/CodeFormer). |
