Skip to main content

Table 1 Different models accuracies on video frames of UCF-101

From: Improved two-stream model for human action recognition

 

Top 1 %

Top 5%

Spatial Stream ConvNet [9]

72.7

 

VGG16

32.1

51.3

Inception V3 [16]

54.55

79.92

VGG16+LSTM (bidirectional)

88.1

96.72

VGG16+LSTM (single directional)

90.81

98.61

  1. The performance includes both top 1 and top 5 accuracies