Fast Newton active appearance models

Kossaifi, J., Tzimiropoulos, G. and Pantic, M. (2014) Fast Newton active appearance models. In: IEEE International Conference on Image Processing, 27 - 30 October 2014, La Defense, Paris.

Full content URL:

Full text not available from this repository.

Item Type:Conference or Workshop contribution (Paper)
Item Status:Live Archive


Active Appearance Models (AAMs) are statistical models of shape and appearance widely used in computer vision to detect landmarks on objects like faces. Fitting an AAM to a new image can be formulated as a non-linear least-squares problem which is typically solved using iterative methods. Owing to its efficiency, Gauss-Newton optimization has been the standard choice over more sophisticated approaches like Newton. In this paper, we show that the AAM problem has structure which can be used to solve efficiently the original Newton problem without any approximations. We then make connections to the original Gauss-Newton algorithm and study experimentally the effect of the additional terms introduced by the Newton formulation on both fitting accuracy and convergence. Based on our derivations, we also propose a combined Newton and Gauss-Newton method which achieves promising fitting and convergence performance. Our findings are validated on two challenging in-the-wild data sets. © 2014 IEEE.

Keywords:Computer vision, Gaussian distribution, Image processing, Image recognition, Inverse problems, Least squares approximations, Newton-Raphson method, Active appearance models, Convergence performance, Gauss-Newton algorithm, Gauss-Newton methods, Gauss-Newton optimization, Inverse compositional image alignments, Levenberg-Marquardt, Nonlinear least squares problems, Iterative methods
Subjects:G Mathematical and Computer Sciences > G450 Multi-media Computing Science
Divisions:College of Science > School of Computer Science
ID Code:21425
Deposited On:04 Mar 2016 14:42

Repository Staff Only: item control page