/openvino_face_recognition

Primary LanguagePythonMIT LicenseMIT

OpenVINO_Human_Face_detection

Using face_detection and facenet for recognizing human identities with the OpenVINO library on an Intel CPU.

Getting Started

Pre-requirements

Install docker before installing the docker container.

  • Tutorial-docker

  • Add docker to sudo group

    sudo groupadd docker
sudo usermod -aG docker $USER
sudo chmod 777 /var/run/docker.sock

Build docker images

sudo chmod 777 ./docker
sudo ./docker/build.sh -m

Run container

sudo ./docker/run.sh

Setting config

{
    "det_model":"models/face_detection/version-RFB-320.xml",
    "landmark_model":"models/facenet/facenet_keras.xml",
    "source":"data/feature/Tom_Hiddleston.jpg",
    "init_folder_path":"data/original/",
    "loop":true,
    "device":"CPU",
    "threshold":0.5
}

Initial face features

python3 init_features.py -c config/config.json

Detection results

python3 facial_recognition.py -c config/config.json

Display

Show inference result

Reference