A local feature-based facial expression recognition system from depth video - Emerging Trends in Image Processing, Computer Vision, and Pattern Recognition - page 348

Table 7
FER Confusion Matrix Using Depth Faces with LDP-PCA-LDA

Expression Anger Happy Sad Surprise Fear Disgust

Anger 97.50% 0 2.50 0 0 0

Happy 2.50 97.50 0 0 0 0

Sad 0 0 97.50 2.50 0 0

Surprise 0 0 0 100 0 0

Fear 0 0 2.50 0 97.50 0

Disgust 0 0 0 0 0 100

Average 98.33

Bold values indicate correct expression recognition and others indicate incorrect recognition.

FIGURE 9 Depth image-based FER performances using different approaches.

5 Concluding Remarks
A depth video-based robust FER system has been proposed in this work using LDP-PCA-
LDA features for facial expression feature extraction and HMM for recognition. The proposed
method was compared with other traditional approaches and the recognition performance
showed its superiority over others. However, the proposed system can be implemented in
many systems such as smart home applications.

References
[1] Uddin MZ, Jehad Sarkar AM. A facial expression recognition system from depth video. In:
Proc. 2014 international conference on image processing, computer vision, & pattern
recognition (IPCV'14), July 21-24, Las Vegas; 2014.
[2] Kim D-S, Jeon I-J, Lee S-Y, Rhee P-K, Chung D-J. Embedded face recognition based
on fast genetic algorithm for intelligent digital photography. IEEE Trans Consum Elec-
tron. 2006;52(3):726-734.
[3] Padget C, Cotrell G. Representation face images for emotion classiication. Cam-
bridge, MA: MIT Press; . Advances in neural information processing systems. 1997;vol. 9.
[4] Mitra S, Acharya T. Gesture recognition: a survey. IEEE Trans Syst Man Cybern C Appl
Rev. 2007;37(3):311-324.
[5] Donato G, Bartlet MS, Hagar JC, Ekman P, Sejnowski TJ. Classifying facial actions.
IEEE Trans Patern Anal Mach Intell. 1999;21(10):974-989.