Skip to main content
Fig. 13 | EURASIP Journal on Image and Video Processing

Fig. 13

From: Inner lips feature extraction based on CLNF with hybrid dynamic template for Cued Speech

Fig. 13

a Speaker’s lips with CLNF original landmarks (note that two red endpoints of the inner lips contour are mistakenly placed). b Six center points are plotted with red stars which are dilated in the vertical scale to form a square in polar coordinates. c The vertical axis presents the ρ value and the horizon axis θ. The six points are repeated three times. d Periodical spline interpolation is realized, and only the period inside two red lines is used to return to Cartesian coordinates. e The whole interpolated inner lips contour

Back to article page