Chi-Squared Distance Metric Learning for Histogram Data

Table 2

Classification errors (%) of NN () using different metrics on the four histogram datasets. Each reported term is an average classification error and a standard deviation over the five runs of cross validation. The minimum classification error of each column is highlighted in bold.

Method	dslr	webcam	amazon	caltech
[15]	22.2 1.8	13.0 1.2	34.3 1.0	58.8 1.1
QCS [15]	25.6 2.7	19.4 1.1	33.9 2.0	57.2 1.2
QCN [15]	27.8 4.1	17.5 2.1	34.5 1.5	56.1 1.2
ITML [15]	25.0 3.0	12.4 1.6	31.6 1.2	52.2 2.1
LMNN [15]	28.9 1.6	15.8 3.0	31.8 1.4	50.9 1.4
GB-LMNN [15]	22.9 2.7	12.4 0.9	29.6 1.7	49.8 1.0
-LMNN [15]	20.6 1.1	8.3 0.9	23.7 0.8	46.5 1.1
CDML