Similarity Measure Learning in Closed-Form Solution for Image Classification

Table 3

Classification performance comparison on the CMU PIE database.

Method	5 Train		10 Train		20 Train		30 Train
	Error (%)	d	Error (%)	d	Error (%)	d	Error (%)	d
PCA	76.6±4.3	334	64.8±4.6	673	48.6±3.8	982	37.9±3.5	1023
RS-2DPCA	44.5±4.1	18	28.3±3.5	18	20.1±2.6	18	9.6±2.2	18
LDA	42.0±3.6	67	29.7±3.7	67	21.5±2.9	67	10.9±3.2	67
LPP	38.0±4.8	67	29.6±3.5	139	20.2±3.3	147	10.8±2.7	86
MFA	36.8±4.4	72	28.2±2.8	69	17.5±2.6	68	9.8±3.0	77
CDA	34.7±3.9	85	23.5±2.5	76	17.3±2.3	79	8.9±2.6	82
CEA	33.5±4.2	241	22.1±2.7	196	14.8±1.9	283	8.4±1.7	129
ISM-GE	32.6±4.1	76	20.7±3.2	73	11.3±2.7	77	6.8±1.6	79
MSE	34.9±4.5	223	25.4±2.5	226	19.0±2.1	230	9.2±1.9	221
CSML	30.4 ± 3.7	201	16.8 ± 2.3	215	9.2 ± 2.2	194	6.1 ± 1.5	192
KCSML	31.8 ± 4.3	211	17.1 ± 2.9	253	6.5 ± 2.4	200	4.3 ± 2.1	203