Attentive 3D-Ghost Module for Dynamic Hand Gesture Recognition with Positive Knowledge Transfer

Table 11

Accuracy comparison of different multimodal fusion-based hand gesture recognition methods on the VIVA dataset.

Method	Fused modalities	Accuracy
HOG + HOG2 [28]	RGB + depth	64.5
CNN:LRN [13]	RGB + depth	74.4
CNN:LRN:HRN [13]	RGB + depth	77.5
C3D [33]	RGB + depth	77.4
I3D [6]	RGB + depth	83.10
MTUT [5]	RGB + depth	86.08
Ours	RGB + depth	86.97

In the experiment results, the best recognition accuracy is in bold.