Project deployment.

The project aims to transform the realm of beauty by developing a sophisticated UNet architecture model where the input represents a woman adorned with makeup, while the output showcases her natural, makeup-free beauty. Furthermore, elevate the project's impact and accessibility by deploying it seamlessly using Flask for web interfacing, TensorFlow Serving for robust serving capabilities, and OpenVINO Serving for optimized inference performance

As an Image Generation problem, we use Structural Similarity Index Measurement (SSIM) to evaluate model's performance. Due to lacking useage free GPU on Colab, we only record the final epoch metric:

• Resnet: 0.91625
• Xception: 0.9351

Test sample 1

Test sample 2

Test sample 3

Clone the project

  git clone https://github.com/TunggTungg/de_makeup.git

Go to the project directory

  cd de_makeup

To deploy this project with TensorFlow Serving

  cd tf_serving
  docker-compose up --build

To deploy this project with OpenVINO Serving

  cd openvino_serving
  docker-compose up --build

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.

MIT