in the version find_best_frame = True
there will be a problem to read the actual code from face alignment.
So if you change the demo.py in line
107 to
fa = face_alignment.FaceAlignment(face_alignment.LandmarksType.TWO_D, flip_input=True, #the 2 is getting to TWO
in the version find_best_frame = True there will be a problem to read the actual code from face alignment.
So if you change the demo.py in line 107 to fa = face_alignment.FaceAlignment(face_alignment.LandmarksType.TWO_D, flip_input=True, #the 2 is getting to TWO
everything will go on fine.