Action Recognition Using Motion Primitives and Probabilistic Edit Distance

In this paper we describe a recognition approach based on the notion of primitives. As opposed to recognizing actions based on temporal trajectories or temporal volumes, primitive-based recognition is based on representing a temporal sequence containing an action by only a few characteristic time instances. The human whereabouts at these instances are extracted by double difference images and represented by four features. In each frame the primitive, if any, that best explains the observed data is identified. This leads to a discrete recognition problem since a video sequence will be converted into a string containing a sequence of symbols, each representing a primitives. After pruning the string a probabilistic Edit Distance classifier is applied to identify which action best describes the pruned string. The approach is evaluated on five one-arm gestures and the recognition rate is 91.3%. This is concluded to be a promising result but also leaves room for further improvements

[1]  Xavier Varona,et al.  aSpaces : Action Spaces for Recognition and Synthesis of Human Actions , 2002, AMDO.

[2]  Aaron F. Bobick,et al.  A State-Based Approach to the Representation and Recognition of Gesture , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Christoph Bregler,et al.  Learning and recognizing human dynamics in video sequences , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[4]  Maja J. Mataric,et al.  Deriving action and behavior primitives from human motion data , 2002, IEEE/RSJ International Conference on Intelligent Robots and Systems.

[5]  Agnès Just,et al.  HMM and IOHMM for the Recognition of Mono- and Bi-Manual 3D Hand Gestures , 2004, BMVC.

[6]  Xiao Li,et al.  Human motion recognition based on neural network , 2005, Proceedings. 2005 International Conference on Communications, Circuits and Systems, 2005..

[7]  Mubarak Shah,et al.  View-Invariant Representation and Recognition of Actions , 2002, International Journal of Computer Vision.

[8]  James W. Davis,et al.  The Recognition of Human Movement Using Temporal Templates , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Vladimir I. Levenshtein,et al.  Binary codes capable of correcting deletions, insertions, and reversals , 1965 .

[10]  Rémi Ronfard,et al.  Motion History Volumes for Free Viewpoint Action Recognition , 2005 .

[11]  Rama Chellappa,et al.  A framework for activity-specific human identification , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[12]  Timothy F. Cootes,et al.  A model of facial behaviour , 2004, Sixth IEEE International Conference on Automatic Face and Gesture Recognition, 2004. Proceedings..

[13]  Aaron F. Bobick,et al.  Recognition of human body motion using phase space constraints , 1995, Proceedings of IEEE International Conference on Computer Vision.

[14]  Agnès Just,et al.  Recognition of isolated complex mono- and bi-manual 3D hand gestures , 2004, Sixth IEEE International Conference on Automatic Face and Gesture Recognition, 2004. Proceedings..

[15]  Thomas B. Moeslund,et al.  Finding Motion Primitives in Human Body Gestures , 2005, Gesture Workshop.

[16]  Jernej Barbic,et al.  Segmenting Motion Capture Data into Distinct Behaviors , 2004, Graphics Interface.

[17]  Mubarak Shah,et al.  Actions sketch: a novel action representation , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).