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

Table 5

The influence of different components of 3DGSAI network on VIVA dataset.

	Testing modality
Method	RGB	Depth
Baseline	80.03	78.97
Baseline + SA	80.32	79.30
Baseline + 3D-Ghost	80.66	79.53
Baseline + 3DGSAI	80.99	79.75

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