Skip to main content
placeholder image

Human action recognition with expandable graphical models

Chapter


Abstract


  • This chapter first presents a brief review of the recent development in human action recognition. In particular, the principle and shortcomings of the conventional Hidden Markov Model (HMM) and its variants are discussed. We then introduce an expandable graphical model that represents the dynamics of human actions using a weighted directed graph, referred to as action graph. Unlike the conventional HMM, the action graph is shared by all actions to be recognized with each action being encoded in one or multiple paths and, thus, can be effectively and efficiently trained from a small number of samples. Furthermore, the action graph is expandable to incorporate new actions without being retrained and compromised. To verify the performance of the proposed expandable graphic model, a system that learns and recognizes human actions from sequences of silhouettes is developed and promising results are obtained. © 2010, IGI Global.

Publication Date


  • 2009

Citation


  • Li, W., Zhang, Z., Liu, Z., & Ogunbona, P. (2009). Human action recognition with expandable graphical models. In Machine Learning for Human Motion Analysis: Theory and Practice (pp. 187-212). doi:10.4018/978-1-60566-900-7.ch010

International Standard Book Number (isbn) 13


  • 9781605669007

Scopus Eid


  • 2-s2.0-84899364610

Web Of Science Accession Number


Book Title


  • Machine Learning for Human Motion Analysis: Theory and Practice

Start Page


  • 187

End Page


  • 212

Abstract


  • This chapter first presents a brief review of the recent development in human action recognition. In particular, the principle and shortcomings of the conventional Hidden Markov Model (HMM) and its variants are discussed. We then introduce an expandable graphical model that represents the dynamics of human actions using a weighted directed graph, referred to as action graph. Unlike the conventional HMM, the action graph is shared by all actions to be recognized with each action being encoded in one or multiple paths and, thus, can be effectively and efficiently trained from a small number of samples. Furthermore, the action graph is expandable to incorporate new actions without being retrained and compromised. To verify the performance of the proposed expandable graphic model, a system that learns and recognizes human actions from sequences of silhouettes is developed and promising results are obtained. © 2010, IGI Global.

Publication Date


  • 2009

Citation


  • Li, W., Zhang, Z., Liu, Z., & Ogunbona, P. (2009). Human action recognition with expandable graphical models. In Machine Learning for Human Motion Analysis: Theory and Practice (pp. 187-212). doi:10.4018/978-1-60566-900-7.ch010

International Standard Book Number (isbn) 13


  • 9781605669007

Scopus Eid


  • 2-s2.0-84899364610

Web Of Science Accession Number


Book Title


  • Machine Learning for Human Motion Analysis: Theory and Practice

Start Page


  • 187

End Page


  • 212