2009 | OriginalPaper | Buchkapitel
View-Invariant Human Action Recognition Using Exemplar-Based Hidden Markov Models
verfasst von : Xiaofei Ji, Honghai Liu
Erschienen in: Intelligent Robotics and Applications
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
An exemplar-based Hidden Markov Model is proposed for human action recognition from any arbitrary viewpoint image sequence. In this framework, human action is modelled as a sequence of body poses (i.e., exemplars) which are represented by a collection of silhouette images. The human actions are recognized by matching the observation image sequence to predefined exemplars, in which the temporal constraints were imposed in the exemplar-based Hidden Markov Model. The proposed method is evaluated in a public dataset and the result shows that it not only reduces computational complexity, but it also is able to accurately recognize human actions using single cameras.