Skip to main content

Table 7 Prediction accuracy comparison on sub-JHMDB (%)

From: Human activity prediction using saliency-aware motion enhancement and weighted LSTM network

Method Accuracy with half videos Accuracy with full videos
PDP [14] 49.2 52.3
Confidence-DT [33] 48.5 50.2
WSF [34] 55.6 73.8
RPAN [16] 64.4 78.6
DRN [22] 71.8 77.3
RGN [23] 66.0 78.0
WLSTM (two modalities) 66.8 78.1