Recognition of action dynamics in fencing using multimodal cues
作者：
Highlights：
• We show that dynamics of action is useful in analysis of similar motion patterns.
• We propose a novel Local Trace Image for action representation and recognition.
• Informative motion descriptors on accelerometric data, skeleton joints and depth maps
• Dataset for analysis of footwork in fencing is introduced (home.agh.edu.pl/~fmal/ffd).
• Fusing data from different modalities permits better recognition accuracy.
摘要
•We show that dynamics of action is useful in analysis of similar motion patterns.•We propose a novel Local Trace Image for action representation and recognition.•Informative motion descriptors on accelerometric data, skeleton joints and depth maps•Dataset for analysis of footwork in fencing is introduced (home.agh.edu.pl/~fmal/ffd).•Fusing data from different modalities permits better recognition accuracy.
论文关键词：Action recognition,Depth maps,Multimodal cues,Motion analysis
论文评审过程：Received 13 January 2017, Revised 24 April 2018, Accepted 25 April 2018, Available online 7 May 2018, Version of Record 21 May 2018.
论文官网地址：https://doi.org/10.1016/j.imavis.2018.04.005
