_How to run easily the face recognition on Raspberry with the help of Docker*
All congrats for M. Marcelo who has done the dev, I've just put that in container.
https://www.hackster.io/mjrobot/real-time-face-recognition-an-end-to-end-project-a10826
X11 externalized, means you have the graphical output from your docker ;)
- Raspberry min 3
- docker installed and running with the pi user
- camera up and running
After have cloned the repository localy (git clone https://github.com/guillain/Rpi-Docker-FacialReco.git
)
and enter in the new directory (cd Rpi-Docker-FacialReco
), execute the following commands:
1/ create your dataset with your face: ./run 01_face_dataset.py
- enter the index corresponding to the person to recognize
- place the faceof this person in front of your camera to take 30 pictures as reference for the recognition
- to run for each face to identify
2/ train and create model: ./run 02_face_training.py
- based on the previous pictures, the model will be trained
3/ run the recognition: ./run 03_face_recognition.py
- to have the name of th person on the picture according to the index, thanks to update the script (sorry)
- run the facial recognition in live
./run
... and jump in the container to play with
- Original post:
- Raspberry: https://www.raspberrypi.org
- Docker: https://www.docker.com
- OpenCV: https://opencv.org
- OpenCV image: https://hub.docker.com/r/sgtwilko/rpi-raspbian-opencv/