论文信息 - To appear in : Advances in Neural Information Processing Systems

To appear in : Advances in Neural Information Processing Systems

Deformable models are an attractive approach to recognizing nonrigid objects which have considerable within class variability. However, there are severe search problems associated with fitting the models to data. We show that by using neural networks to provide better starting points, the search time can be significantly reduced. The method is demonstrated on a character recognition task.

Geoffrey E. Hinton | Christopher K. I. Williams | Michael Revow | M. Revow

[1] William Grimson,et al. Object recognition by computer - the role of geometric constraints , 1991 .

[2] Loren W. Nolte,et al. Performance of the hough transform and its relationship to statistical signal detection theory , 1988, Comput. Vis. Graph. Image Process..

[3] Geoffrey E. Hinton,et al. Discovering Viewpoint-Invariant Relationships That Characterize Objects , 1990, NIPS.

[4] Geoffrey E. Hinton,et al. Adaptive Elastic Models for Hand-Printed Character Recognition , 1991, NIPS.

[5] David J. C. MacKay,et al. Bayesian Interpolation , 1992, Neural Computation.

[6] Dana H. Ballard,et al. Generalizing the Hough transform to detect arbitrary shapes , 1981, Pattern Recognit..

[7] Ulf Grenander,et al. Hands: A Pattern Theoretic Study of Biological Shapes , 1990 .

[8] Geoffrey E. Hinton,et al. Adaptive Mixtures of Local Experts , 1991, Neural Computation.

[9] Geoffrey E. Hinton,et al. Combining deformable models and neural networks for handprinted digit recognition , 1994 .

[10] Geoffrey E. Hinton,et al. Combining two methods of recognizing hand-printed digits , 1992 .

[11] Christopher K. I. Williams,et al. Using mixtures of deformable models to capture variations in hand printed digits , .

[12] Tomaso Poggio,et al. Example Based Image Analysis and Synthesis , 1993 .