ISSN : 1796-2048
Volume : 2    Issue : 5    Date : September 2007

Hierarchical Model-Based Activity Recognition With Automatic Low-Level State Discovery
Justin Muncaster and Yunqian Ma
Page(s): 66-76
Full Text:
PDF (585 KB)

Activity recognition in video streams is increasingly important for both the computer vision and
artificial intelligence communities. Activity recognition has many applications in security and video
surveillance. Ultimately in such applications one wishes to recognize complex activities, which can
be viewed as combination of simple activities. In this paper, we present a general framework of a
Dlevel dynamic Bayesian network to perform complex activity recognition. The levels of the network
are constrained to enforce state hierarchy while the Dth level models the duration of simplest event.
Moreover, in this paper we propose to use the deterministic annealing clustering method to
automatically define the simple activities, which corresponds to the low level states of observable
levels in a Dynamic Bayesian Networks. We used real data sets for experiments. The experimental
results show the effectiveness of our proposed method.

Index Terms
Activity Recognition, Dynamic Bayesian Networks, Deterministic Annealing, Video Surveillance