Parameter estimation of two dimensional component Gaussian mixtures

Multiresolution Gaussian Mixture Models (MGMM) can be used to represent image and video data in video annotation and retrieval. Preliminary experiments were carried out to estimate the model parameters for two-dimensional data. An iterative algorithm similar to Expectation-Maximisation (EM) is used to estimate the model parameters. The suitability of Akaike's Information Criterion (AIC) as a measure of model fit is also evaluated. AIC was successful for most of the synthetic data sets used in the experiments, however further work is required to develop a more consistent criterion for model fit.

[1]  P. Smith.  On Akaike's model-fit criterion , 1987 .

[2]  H. Akaike.  A new look at the statistical model identification , 1974 .

[3]  Zucchini,et al.  An Introduction to Model Selection. , 2000, Journal of mathematical psychology.

[4]  R. Jaszczak,et al.  Parameter estimation of finite mixtures using the EM algorithm and information criteria with application to medical image processing , 1992 .

[5]  Roland Wilson,et al.  MGMM: multiresolution Gaussian mixture models for computer vision , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[6]  P. McKenzie,et al.  Selecting the optimal number of components for a Gaussian mixture model , 1994, Proceedings of 1994 IEEE International Symposium on Information Theory.

[7]  Richard A. Johnson,et al.  Applied Multivariate Statistical Analysis , 1983 .

[8]  Tad J. Ulrych,et al.  Information theory measures with application to model identification , 1986, IEEE Trans. Acoust. Speech Signal Process..

[9]  A. F. Smith,et al.  Statistical analysis of finite mixture distributions , 1986 .

[10]  Anil K. Jain,et al.  Unsupervised selection and estimation of finite mixture models , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.