Abstract. We present a method for learning characteristic motion patterns of mobile agents. The method works on two levels. On the first level, it uses the expectation-maximization...
The Gaussian mixture model is a powerful statistical tool in data modeling and analysis. Generally, the EM algorithm is utilized to learn the parameters of the Gaussian mixture. Ho...
This paper proposes an interpolating extension to hidden Markov models (HMMs), which allows more accurate modeling of natural sounds sources. The model is able to produce observat...
We present a two-layer generative model for sport video mining that is composed of a two-layer observation model. The first layer is the Gaussian mixture model (GMM) using framew...