Variational Bayesian image modelling

ABSTRACT

We present a variational Bayesian framework for performing inference, density estimation and model selection in a special class of graphical models---Hidden Markov Random Fields (HMRFs). HMRFs are particularly well suited to image modelling and in this paper, we apply them to the problem of image segmentation. Unfortunately, HMRFs are notoriously hard to train and use because the exact inference problems they create are intractable. Our main contribution is to introduce an efficient variational approach for performing approximate inference of the Bayesian formulation of HMRFs, which we can then apply to the density estimation and model selection problems that arise when learning image models from data. With this variational approach, we can conveniently tackle the problem of image segmentation. We present experimental results which show that our technique outperforms recent HMRF-based segmentation methods on real world images.

REFERENCES

• 1: Ackley, D., Hinton, G., & Sejnowski, T. (1985). A learning algorithm for Boltzmann machine. Cognitive Science, 9, 147--169.
• 2: Attias, H. (2000). A variational Bayesian framework for graphical models. Advances in Neural Information Processing Systems 12 (pp. 209--215).
• 3: Beal, M., & Ghahramani, Z. (2003). The variational Bayesian EM algorithm for incomplete data: with application to scoring graphical model structures. Bayesian Statistics 7 (pp. 453--464).
• 4: Besag, J. (1986). On the statistical analysis of dirty pictures (with discussions). Journal of the Royal Statistical Society, Series B, 48, 259--302.
• 5: Florence Forbes , Nathalie Peyrard, Hidden Markov Random Field Model Selection Criteria Based on Mean Field-Like Approximations, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.25 n.9, p.1089-1101, September 2003 [doi>10.1109/TPAMI.2003.1227985]
• 6: F. Heitz , P. Bouthemy, Multimodal Estimation of Discontinuous Optical Flow using Markov Random Fields, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.15 n.12, p.1217-1232, December 1993 [doi>10.1109/34.250841 ]
• 7: Michael I. Jordan , Zoubin Ghahramani , Tommi S. Jaakkola , Lawrence K. Saul, An Introduction to Variational Methods for Graphical Models, Machine Learning, v.37 n.2, p.183-233, Nov.1.1999 [doi>10.1023/A:1007665907178 ]
• 8: David John Cameron Mackay, Bayesian methods for adaptive models, California Institute of Technology, Pasadena, CA, 1992
• 9: Radford M. Neal , Geoffrey E. Hinton, A view of the EM algorithm that justifies incremental, sparse, and other variants, Learning in graphical models, MIT Press, Cambridge, MA, 1999
• 10: Opper, M., & Saad, D. (Eds.). (2001). Advanced mean field methods: Theory and practice. the MIT press.
• 11: O'Ruanaidh, J., & Fitzgerald, W. (1996). Numerical Bayesian methods applied to signal processing. Statistics and Computing. Springer.
• 12: Derek C. Stanford , Adrian E. Raftery, Approximate Bayes Factors for Image Segmentation: The Pseudolikelihood Information Criterion (PLIC), IEEE Transactions on Pattern Analysis and Machine Intelligence, v.24 n.11, p.1517-1520, November 2002 [doi>10.1109/TPAMI.2002.1046170 ]
• 13: Zhuowen Tu , Song-Chun Zhu, Image Segmentation by Data-Driven Markov Chain Monte Carlo, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.24 n.5, p.657-673, May 2002 [doi>10.1109/34.1000239 ]
• 14: Wainwright, M. J., & Jordan, M. I. (2003). Graphical models, exponential families, and variational inference (Technical Report 649). UC Berkeley, Dept. of Statistics.
• 15: Jonathan S. Yedidia , William T. Freeman , Yair Weiss, Understanding belief propagation and its generalizations, Exploring artificial intelligence in the new millennium, Morgan Kaufmann Publishers Inc., San Francisco, CA, 2003
• 16: Zhang, J. (1992). The mean field theory in EM procedures for Markov random fields. IEEE Transaction on Signal Processing, 40, 2570--2583.
• 17: Zhang, P. (1993). Model selection via multifold cross validation. Annals of Statistics, 21, 299--313.