A general non-linear method for modelling shape and locating image objects
- 1 January 1996
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 4 (10514651) , 266-270 vol.4
- https://doi.org/10.1109/icpr.1996.547428
Abstract
Objects of the same class often exhibit variation in shape. This shape variation has previously been modelled by means of point distribution models (PDMs) in which there is a linear relationship between a set of shape parameters and the positions of points on the shape. Here we present a new form of PDM, which uses a multilayer perceptron (MLP) to carry out nonlinear principal component analysis. We demonstrate that MLP-PDMs can model the shape variability in classes of object for which the linear model fails. We describe the use of MLP-PDMs in image search and present quantitative results for a practical application (face recognition), demonstrating the ability to locate image structures accurately starting from a very poor initial approximation to their pose and shape.Keywords
This publication has 10 references indexed in Scilit:
- A unified approach to coding and interpreting face imagesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Nonlinear manifold learning for visual speech recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Non-linear point distribution modelling using a multi-layer perceptronImage and Vision Computing, 1997
- Non-linear generalization of point distribution models using polynomial regressionImage and Vision Computing, 1995
- Active Shape Models-Their Training and ApplicationComputer Vision and Image Understanding, 1995
- Automated Pivot Location for the Cartesian-Polar Hybrid Point Distribution Model.Published by British Machine Vision Association and Society for Pattern Recognition ,1995
- Flexible 3D Models from Uncalibrated Cameras.Published by British Machine Vision Association and Society for Pattern Recognition ,1995
- Active Shape Models: Evaluation of a Multi-Resolution Method for Improving Image SearchPublished by British Machine Vision Association and Society for Pattern Recognition ,1994
- Nonlinear principal component analysis using autoassociative neural networksAIChE Journal, 1991
- An Algorithm for Finding Intrinsic Dimensionality of DataIEEE Transactions on Computers, 1971