A Nonlinear Discriminative Approach to AAM Fitting

The Active Appearance Model (AAM) is a powerful generative method for modeling and registering deformable visual objects. Most methods for AAM fitting utilize a linear parameter update model in an iterative framework. Despite its popularity, the scope of this approach is severely restricted, both in fitting accuracy and capture range, due to the simplicity of the linear update models used. In this paper, we present an new AAM fitting formulation, which utilizes a nonlinear update model. To motivate our approach, we compare its performance against two popular fitting methods on two publicly available face databases, in which this formulation boasts significant performance improvements.

[1]  Simon Baker,et al.  Equivalence and efficiency of image alignment algorithms , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[2]  Roland Göcke,et al.  Iterative Error Bound Minimisation for AAM Alignment , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[3]  Ralph Gross,et al.  Generic vs. person specific active appearance models , 2005, Image Vis. Comput..

[4]  Ankur Agarwal,et al.  Recovering 3D human pose from monocular images , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[6]  Björn Stenger,et al.  Multivariate Relevance Vector Machines for Tracking , 2006, ECCV.

[7]  Timothy F. Cootes,et al.  Active Appearance Models , 1998, ECCV.

[8]  Rainer Lienhart,et al.  An extended set of Haar-like features for rapid object detection , 2002, Proceedings. International Conference on Image Processing.

[9]  Simon Baker,et al.  Active Appearance Models Revisited , 2004, International Journal of Computer Vision.

[10]  Mikkel B. Stegmann,et al.  Fast Registration of Cardiac Perfusion MRI , 2003 .

[11]  Simon Baker,et al.  Lucas-Kanade 20 Years On: A Unifying Framework , 2004, International Journal of Computer Vision.

[12]  Andrew Blake,et al.  A sparse probabilistic learning algorithm for real-time tracking , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[13]  Dorin Comaniciu,et al.  Image based regression using boosting method , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[14]  Avinash C. Kak,et al.  Accurate 3D Tracking of Rigid Objects with Occlusion Using Active Appearance Models , 2005, 2005 Seventh IEEE Workshops on Applications of Computer Vision (WACV/MOTION'05) - Volume 1.

[15]  Horst Bischof,et al.  Fast Active Appearance Model Search Using Canonical Correlation Analysis , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Mikkel B. Stegmann,et al.  The IMM Face Database, An Annotated Dataset of 240 Face Images , 2004 .

[17]  Jiri Matas,et al.  XM2VTSDB: The Extended M2VTS Database , 1999 .

[18]  Stan Z. Li,et al.  Direct appearance models , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[19]  Timothy F. Cootes,et al.  Interpreting face images using active appearance models , 1998, Proceedings Third IEEE International Conference on Automatic Face and Gesture Recognition.

[20]  Aziz Umit Batur,et al.  Adaptive active appearance models , 2005, IEEE Transactions on Image Processing.

[21]  P. Kittipanya-ngam,et al.  The effect of texture representations on AAM performance , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[22]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[23]  Rasmus Larsen,et al.  Multi-band modelling of appearance , 2003, Image Vis. Comput..

[24]  L. Petersson,et al.  Response Binning: Improved Weak Classifiers for Boosting , 2006, 2006 IEEE Intelligent Vehicles Symposium.