Skip to main content
Fig. 15 | EURASIP Journal on Image and Video Processing

Fig. 15

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

Fig. 15

Red curve: B parameter (expressed in pixels) determined by the ground truth value. Blue curve: B values by original CLNF landmarks. Black curve: B parameter by proposed method. Green curve: errors of original CLNF. Magenta curve: errors of Modified CLNF. The figure shows the result of three speakers from the top to the bottom. The blue short lines at the top are the segmentation for 50 words. The cyan line is the RMSE value

Back to article page