JordanPeltier

+1 for this feature :)

aspect_ratios = [2] with Max_size set to None means you will have the ratios 1.0, 2.0 and 1/2.0. So actually you will have 3 priors

Same question for the variance

Same here. Training is happening but very slow because of this warning

Hi, Actually I found the solution. You have to use the method "extended" in order to remove the non-extended fingers of the list. I probably don't have the same SDK...

The problem is that the facial landmarks are very good. And I tried with different angles. Let me show you :    

Hi, So it may be my set up : > > > cv2.**version** > > > '3.1.0' > > > dlib.**version** > > > '18.18.100'

Ok thanks for this reply !

@dougsouza Thanks for considering my issue. I attach what i get with bigger face.    

@dougsouza Do I have to preprocess the picture before using the frontalize code ? Like face detection, resize or something?