Skip to main content

Table 4 Comparison of the overall performance of our method and the trajectory-based methods

From: Action recognition using length-variable edge trajectory and spatio-temporal motion skeleton descriptor

Methods

UCF Sports (%)

YouTube (%)

HMDB51 (%)

Wang et al. [23]

88.0

84.1

46.6

Seo et al. [16]

–

–

57.8

Yi et al. [30]

90.0

–

–

Wang et al. [39]

92.0

88.9

54.3

Peng et al. [40]

–

87.6

51.8

Cho et al. [41]

89.7

86.1

–

Ours

92.8

89.6

58.2

  1. The italicized values indicate the best HAR performance among the three datasets