Fig. 8From: Inner lips feature extraction based on CLNF with hybrid dynamic template for Cued SpeechTwenty black lips landmarks are given by the CLNF with 8 inner lips landmarks. The green point A is the middle outer lower lips position obtained by the HD-CTM and the green point B is the middle inner lower lips position by subtracting the VLLT from the position of A. Two orange points are the left and right key landmarks determined by the third stepBack to article page