A bag of words approach to subject specific 3D human pose interaction classification with random decision forests

• Published in: Graphical models Vol. 76; no. 3; pp. 162 - 171
• Main Authors: Deng, Jingjing, Xie, Xianghua, Daubney, Ben
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Inc 01.05.2014; Elsevier
• Subjects: Action recognition; Applied sciences; Artificial intelligence; Bag of words; Classification; Computer science; control theory; systems; Computer systems and distributed systems. User interface; Conversation; Data processing. List processing. Character string processing; Decision trees; Exact sciences and technology; Human interaction; Human pose; Low level; Memory organisation. Data processing; Pattern recognition. Digital image processing. Computational geometry; Random forests; Recognition; Software; Temporal distribution; Three dimensional; Visual; Action recognition; Bag of words; Human pose; Human interaction; Random forests; Action; Event detection; Speed measurement; Conversation; Behavioral analysis; User interface; Classification; Gesture; Displacement measurement; Microdisplacement; Learning algorithm; Computer vision; Motion estimation; Information retrieval; Random decision forests; Interactive system; Wavelet transformation; Supervised learning; Social interaction; Scene analysis; Tridimensional image

In this work, we investigate whether it is possible to distinguish conversational interactions from observing human motion alone, in particular gestures in 3D. We adopt Kinect sensors to obtain 3D displacement and velocity measurements, followed by wavelet decomposition to extract low level temporal...