Multiparameter Space Decision Voting and Fusion Features for Facial Expression Recognition

Table 3

Comparison of the ER accuracy (%) of various features on the CK + dataset.

Method	AN	CO	DI	FE	HA	SA	SU	ER accuracy (%)
LBP + SVM	83.22	100.00	95.12	93.80	100.00	94.18	92.03	94.05
HOG + SVM	82.22	95.00	91.36	92.00	100.00	100.00	90.00	92.94
Gabor + SVM	83.38	95.64	83.03	60.00	89.78	57.42	95.22	85.76
The proposed method	93.39	100.00	94.29	100.00	95.71	98.75	98.75	98.21