Name
Abstract | ||
|---|---|---|
Standard hidden Markov model (HMM) and the more general dynamic Bayesian network (DBN) models assume stationarity of state transition distribution. However, this assumption does not hold for many real life events of interest. In this paper, we propose a new time sequence model that extends HMM to time varying scenario. The time varying property is realized in our model by explicitly allowing the change of state transition density as the time spent in a particular state passes by. Rather than keeping transition densities at different time spots independent of each other, we exploit their temporal correlation by applying a hierarchical Dirichlet prior. This leads to a more robust time varying model, especially when training data are scarce. We also employ Markov chain Monte Carlo (MCMC) sampling in learning the MAP estimate of time varying parameters, with a transition kernel incorporating linear optimization. The proposed model is applied to recognizing real video events, and is shown to outperform existing HMM-based methods. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1109/ICASSP.2009.4959945 | ICASSP |
Keywords | Field | DocType |
varying model,time varying hidden markov model,transition kernel,transition density,dynamic bayesian network model,hmm,image recognition,markov chain monte carlo sampling,hierarchical dirichlet prior,mcmc,robust time,event recognition,state transition distribution,image sequences,monte carlo methods,hidden markov model,state transition density,different time spot,time varying property,time varying,time sequence model,hidden markov models,new time sequence model,video sequence,linear optimization,contamination,kernel,dirichlet prior,markov chain monte carlo,bayesian methods,information processing,computational modeling,robustness,data mining,dynamic bayesian network,probability density function,state transition | Monte Carlo method,Pattern recognition,Markov chain Monte Carlo,Markov model,Computer science,Artificial intelligence,Dirichlet distribution,Maximum a posteriori estimation,Hidden Markov model,Hidden semi-Markov model,Dynamic Bayesian network | Conference |
ISSN | ISBN | Citations |
1520-6149 E-ISBN : 978-1-4244-2354-5 | 978-1-4244-2354-5 | 5 |
PageRank | References | Authors |
0.44 | 2 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Zhaowen Wang | 1 | 1063 | 40.64 |
| Ercan E. Kuruoglu | 2 | 505 | 61.27 |
| Xiaokang Yang | 3 | 3581 | 238.09 |
| Yi Xu | 4 | 1757 | 177.61 |
| Yu Song | 5 | 356 | 52.74 |