Real-time frontal face detection via SOD Realnets (~ 5 ms on HD video).
This is the built-in realnet embedded face detection model. There is no need to bring an independent model in order to detect faces at real-time.
All you need, is to register the model hex array defined in "sod_face_realnet.h" via `sod_realnet_load_model_from_mem()` and use the same C/C++ as for detecting faces via Realnets.
This feature is available in the commercial version of the library. You can obtain your commercial license from https://pixlab.io/downloads.
Advantages includes:
* Multi-core CPU support for all platforms - Up to 3 ~ 10 times faster processing speed.
* Built-in (C Code), high performance RealNets frontal face detector.
* 75 days of integration & technical assistance.
* Royalty-free commercial licenses without any GPL restrictions.
* Application source code stays private.
* Skeletonization is useful when we are interested not in the size of the pattern but rather
* in the relative position of the strokes in the pattern (Character Recognition,
* X, Y Chromosome Recognition, etc.).
* The target image must be binary (i.e. images whose pixels have only two possible intensity
* value mostly black or white). You can obtain a binary image via sod_canny_edge_image(),
* sod_otsu_binarize_image(), sod_binarize_image() or sod_threshold_image().
Sean McBride
phani
Symisc Systems