Robust human activity recognition from depth video using spatiotemporal multi-fused features
作者：
Highlights：
• We propose novel multi-fused features for online HAR system.
• They are skeleton joint features including joint features DT, DK, M, ⊖ and shape feature HOG-DDS.
• The HAR systems recognize human activities from continuous sequences of depth map.
• It trains the hidden Markov model (HMM) with the code vectors of the multi-fused features.
• It outperforms the state-of-the-art HAR methods in terms of recognition accuracy.
摘要
Highlights•We propose novel multi-fused features for online HAR system.•They are skeleton joint features including joint features DT, DK, M, ⊖ and shape feature HOG-DDS.•The HAR systems recognize human activities from continuous sequences of depth map.•It trains the hidden Markov model (HMM) with the code vectors of the multi-fused features.•It outperforms the state-of-the-art HAR methods in terms of recognition accuracy.
论文关键词：Human activity recognition,Depth silhouette,Skeleton joint extraction,Spatiotemporal multi-fused feature extraction,Hidden Markov model,Forward spotting scheme
论文评审过程：Received 18 May 2015, Revised 16 May 2016, Accepted 3 August 2016, Available online 4 August 2016, Version of Record 16 August 2016.
论文官网地址：https://doi.org/10.1016/j.patcog.2016.08.003
