Face recognition with Python3 using OpenSource library face_recognition
A simple goto Face Recognition
It accepts both images and videos for training. So you can just:
-
Create directory and put all images and video belonging to a particular person
-
Rename directory to the respective person name
-
Repeat with the remaining people
-
Note:
- An image or a video should contain only the face of the respective person
- Optional: Move all directories with images to data/raw_data/
Install dependencies
# install cmake
sudo apt-get install build-essential cmake pkg-config
# Install requirements. Dlib installation will take time (Building from src).
sudo pip3 install -r requirements.txt
Now Prepare Data (without GPU it took me 20 min to encode 3000 images)
# if all directories with images has been moved to data/raw_data/
python3 prepare_data.py
# Otherwise
python3 prepare_data.py -p PATH_TO_DATA_DIRECTORIES
Once completed. We can start testing
# face_recognition using image
python3 predict_face.py -i IMAGE_PATH
# face_recognition using video
python3 predict_face.py -v VIDEO_PATH