Multiparameter Space Decision Voting and Fusion Features for Facial Expression Recognition

Table 7

Comparisons with various feature extraction approaches on the CK+ dataset.

Ref.	Feature	Class	Evaluation	Classification	ER accuracy (%)
2012 [70]	Common + specific patch	6	10-fold	SVM	88.25
2013 [71]	LBP + geometric feature	6	5-fold	SVM	89.56
2014 [45]	LPQ + PHOG	7	7-fold	SVM	93.21
2017 [72]	Data-driven	7	Leave-one-out	SVM	94.81
The proposed method	Fusion feature	7	10-fold	Decision voting	98.21