Face detectors and tracking

Project based on the following articles (entirely recommend for anyone generally interested in the topics):

• Face Detection – OpenCV, Dlib and Deep Learning ( C++ / Python ) https://www.learnopencv.com/face-detection-opencv-dlib-and-deep-learning-c-python/

• Object tracking with dlib https://www.pyimagesearch.com/2018/10/22/object-tracking-with-dlib/

To run the current code (assuming you have installed dlib and opencv for python3.6, recommend to use a virtual environment):

usage: face_detector.py [-h]

[--detect [{no_detection,opencv_haar,opencv_dnn,dlib_hog,dlib_dnn}]]

[--tracking] [--resize_width RESIZE_WIDTH]

[--opencv_dnn_type {TF,CAFFE}]

[--tracking_freq TRACKING_FREQ]

optional arguments:

-h, --help show this help message and exit

--detect [{no_detection,opencv_haar,opencv_dnn,dlib_hog,dlib_dnn}] Options for face detection: ['no_detection', 'opencv_haar', 'opencv_dnn', 'dlib_hog', 'dlib_dnn']

--tracking If specified, tracking is on

--resize_width RESIZE_WIDTH Resize input frame. Maintains aspect ratio.

--opencv_dnn_type {TF,CAFFE}

--tracking_freq TRACKING_FREQ Tracking frequency.