MLP or distance?

[brealisty]

thank you for your work. have your ever compare with other way for classfing the faceID, such as feature distance not to train a MLP classifer.
As we know, MLP classifer is diffcult to expand when there are some new faceIDs and need more face datas. why you choose this way, is this get a better accuracy? how to make choose?

[brealisty]

and in face_detector.py code, the function precess align face_image(it doesn't seem to), but I test some image, get this result: