Abstract

In this paper we propose a novel inhomogeneous Gibbs model by the minimax entropy principle, and apply it to face modeling. The maximum entropy principle generalizes the statistical properties of the observed samples and results in the Gibbs distribution, while the minimum entropy principle makes the learnt distribution close to the observed one. To capture the fine details of a face, an inhomogeneous Gibbs model is derived to learn the local statistics of facial feature points. To alleviate the high dimensionality problem of face models, we propose to learn the distribution in a subspace reduced by principal component analysis or PCA. We demonstrate that our model effectively captures important and subtle non-Gaussian face patterns and efficiently generates good face models. 1.

Citations

3909	Neural Networks for Pattern Recognition - Bishop - 1995
2223	A.: Eigenfaces for recognition - TURK, PENTLAND - 1991
764	Neural network-based face detection - Rowley, Baluja, et al. - 1996
382	T.: Learning low-level vision - FREEMAN, PASZTOR, et al.
335	Mardia K: Statistical Shape Analysis - Dryden - 1998
228	Active shape models—their training and application, Computer Vision and Image Understanding 61 - Cootes, Taylor, et al. - 1995
200	Statistical models of appearance for computer vision. tech. report. university of manchester - Cootes, Taylor - 2001
165	Minimax entropy principle and its applications to texture modeling - Zhu, Wu, et al. - 1997
111	Structural image restoration through deformable template - Amit, Grenander, et al.
98	Deformable templates for face recognition - Yuille - 1991
66	Joint Induction of Shape Features and Tree Classifiers - Amit, Geman, et al. - 1997
64	Size and shape spaces for landmark data in two dimensions (with discussion - Bookstein - 1986
48	Embedding Gestalt laws in Markov random fields - Zhu - 1999
39	Hallucinating faces - Baker, Kanade - 1999
35	Learning to estimate scenes from images - Freeman, Pasztor - 1999
4	Recognition and Neural Networks - Pattern - 1996