mini-face 0.0.11

A minimalist Python wrapper for OpenFace library

Mini-face

Installation with pip

Requirements: python 3.12. Install the package with a following command:

   python3 -m pip install mini-face

Installation from source

Requirements: cmake, vcpkg, python 3.12.
Recommended to use a virtual environment.

1. Install the scikit-build module:

pip install scikit-build

2. Download and install the repository:

git clone https://github.com/child-lab-uj/gaze-tracking
cd gaze-tracking
pip install .

Short description of the project structure:

• /src/ - source code - OpenFace source, wrappers and Python bindings
• /model/ - Model data from the official OpenFace release
• /AU_predictors/ - Model data from the official OpenFace release

Python usage

Below is a minimalist example of using the module from Python code:

   import cv2
   import mini_face.api

   if __name__ == "__main__":
      image = cv2.imread("test_image.jpg")
      roi = (214.467, 96.9926, 110.877, 117.08)

      extractor = mini_face.api.GazeExtractor()

      extractor.estimate_camera_calibration(image)

      print(extractor.detect_gaze(image, 0, roi))