Scaling Human-Object Interaction Recognition in the Video through Zero-Shot Learning

Table 6

The impact of the RNNs (LSTM/GRU) on the proposed system’s performance. Training and testing are performed on all data.

Method	mAP (%)
	LSTM	GRU
2Stream + WE - VLAD	22.45	22.20
3Stream + WE - VLAD	23.76	23.52
2Stream + WE + VLAD(rnn)	23.64	23.43
3Stream + WE + VLAD(rnn)	24.73	24.58