Continuous Regression for Non-rigid Image Alignment

  • Enrique Sánchez-Lozano
  • Fernando De la Torre
  • Daniel González-Jiménez
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7578)


Parameterized Appearance Models (PAMs) such as Active Appearance Models (AAMs), Morphable Models and Boosted Appearance Models have been extensively used for face alignment. Broadly speaking, PAMs methods can be classified into generative and discriminative. Discriminative methods learn a mapping between appearance features and motion parameters (rigid and non-rigid). While discriminative approaches have some advantages (e.g., feature weighting, improved generalization), they suffer from two major drawbacks: (1) they need large amounts of perturbed samples to train a regressor or classifier, making the training process computationally expensive in space and time. (2) It is not practical to uniformly sample the space of motion parameters. In practice, there are regions of the motion space that are more densely sampled than others, resulting in biased models and lack of generalization. To solve these problems, this paper proposes a computationally efficient continuous regressor that does not require the sampling stage. Experiments on real data show the improvement in memory and time requirements to train a discriminative appearance model, as well as improved generalization.


Training Image Motion Parameter Canonical Correlation Analysis Appearance Model Active Appearance Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Cootes, T.F., Edwards, G.J., Taylor, C.J.: Active Appearance Models. PAMI 23(6), 681–685 (2001)CrossRefGoogle Scholar
  2. 2.
    Hao, W., Liu, X., Doretto, G.: Face alignment via boosted ranking model. In: CVPR (2008)Google Scholar
  3. 3.
    Igual, L., De la Torre, F.: Continuous Procrustes Analysis to Learn 2D Shape Models from 3D Objects. In: 3rd Workshop on Non-Rigid Shape and Deformable Image Alignment, in Conjunction with CVPR, pp. 17–22 (2010)Google Scholar
  4. 4.
    Nguyen, M.H., De la Torre, F.: Metric Learning for Image Alignment. Int’l Journal of Computer Vision 88(1), 69–84 (2010)CrossRefGoogle Scholar
  5. 5.
    Jones, M.J., Poggio, T.: Multidimensional Morphable models. In: ICCV (1998)Google Scholar
  6. 6.
    Black, M.J., Jepson, A.D.: Eigentracking: Robust Matching and Tracking of Objects of Articulated Objects Using a View-Based Representation. Int’l Journal of Computer Vision 26(1), 63–64 (1998)CrossRefGoogle Scholar
  7. 7.
    De la Torre, F.: A least-squares framework for Component Analysis. PAMI 34(6), 1041–1055 (2012)CrossRefGoogle Scholar
  8. 8.
    Huang, D., Storer, M., De la Torre, F., Bischof, H.: Supervised Local Subspace Learning for Continuous Head Pose Estimation. In: CVPR (2011)Google Scholar
  9. 9.
    Donner, R., Reiter, M., Langs, G., Pelloscheck, P., Bischof, H.: Fast active appearance model search using canonical correlation analysis. PAMI 28(10), 1690–1694 (2006)CrossRefGoogle Scholar
  10. 10.
    Rivera, S., Martinez, A.M.: Learning Deformable Shape Manifolds. Pattern Recognition 45(4), 1792–1801 (2012)CrossRefGoogle Scholar
  11. 11.
    Valstar, M., Martínez, B., Binefa, X., Pantic, M.: Facial Point Detection using Boosted Regression and Graph Models. In: CVPR (2010)Google Scholar
  12. 12.
    Tian, Y., Narasimhan, S.: Globally Optimal Estimation of Nonrigid Image Distortion. Int’l Journal of Computer Vision (2010)Google Scholar
  13. 13.
    Liu, X.: Discriminative Face Alignment. PAMI 31(11), 1941–1954 (2009)CrossRefGoogle Scholar
  14. 14.
    Liu, X.: Optimal gradient pursuit for face alignment. In: AFGR and Workshops (2011)Google Scholar
  15. 15.
    De la Torre, F., Nguyen, M.H.: Parameterized Kernel Principal Component Analysis: Theory and Applications to Supervised and Unsupervised Image Alignment. In: CVPR (2008)Google Scholar
  16. 16.
    Tresadern, P., Sauer, P., Cootes, T.F.: Additive Update Predictors in Active Appearance Models. In: BMVC (2010)Google Scholar
  17. 17.
    Cootes, T.F., Taylor, C.J.: Statistical Models of Appearance for Computer Vision. Online Technical Report (2004),
  18. 18.
    Sauer, P., Cootes, T.F., Taylor, C.J.: Accurate Regression Procedures for Active Appearance Models. In: BMVC (2011)Google Scholar
  19. 19.
    Cootes, T.F., Edwards, G.J., Taylor, C.J.: A Comparative Evaluation of Active Appearance Model Algorithms. In: BMVC, vol. 2 (1998)Google Scholar
  20. 20.
    Saragih, J., Göcke, R.: Learning AAM fitting through simulation. Int’l Journal on Pattern Recognition 42(11), 2628–2636 (2009)zbMATHCrossRefGoogle Scholar
  21. 21.
    Belhumeur, P.N., Jacobs, D.W., Kriegman, D.J., Kumar, N.: Localizing Parts of Faces Using a Consensus of Exemplars. In: CVPR (2011)Google Scholar
  22. 22.
    Baker, S., Matthews, I.: Lucas-Kanade 20 Years On: A Unifying Framework. Int’l Journal on Computer Vision 56(3), 221–255 (2004)CrossRefGoogle Scholar
  23. 23.
    Matthews, I., Baker, S.: Active Appearance Models Revisited. Int’l Journal on Computer Vision 60(2), 135–164 (2004)CrossRefGoogle Scholar
  24. 24.
    Lucas, B., Kanade, T.: An Iterative image registration technique with an application to stereo vision. In: IJCAI 1981 (1981)Google Scholar
  25. 25.
    Ramsay, J., Silverman, B.: Functional Data Analysis. Springer (1997)Google Scholar
  26. 26.
    Levin, A., Shashua, A.: Principal Component Analysis over Continuous Subspaces and Intersection of Half-Spaces. In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds.) ECCV 2002, Part III. LNCS, vol. 2352, pp. 635–650. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  27. 27.
    Gross, R., Matthews, I., Cohn, J.F., Kanade, T., Baker, S.: Multi-PIE. In: AFGR (2008)Google Scholar
  28. 28.
    Jesorsky, O., Kirchberg, K.J., Frischholz, R.W.: Robust Face Detection Using the Hausdorff Distance. In: Bigun, J., Smeraldi, F. (eds.) AVBPA 2001. LNCS, vol. 2091, pp. 90–95. Springer, Heidelberg (2001)CrossRefGoogle Scholar
  29. 29.
    Saragih, J., Lucey, S., Cohn, J.F.: Deformable Model Fitting by Regularized Landmark Mean-Shift. Int’l Journal of Computer Vision 91(2), 200–215 (2011)MathSciNetzbMATHCrossRefGoogle Scholar
  30. 30.
    Zhu, X., Ramanan, D.: Face detection, pose estimation and landmark localization in the wild. In: CVPR (2012)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Enrique Sánchez-Lozano
    • 1
  • Fernando De la Torre
    • 2
  • Daniel González-Jiménez
    • 1
  1. 1.Multimodal Information AreaVigo, PontevedraSpain
  2. 2.Robotics InstituteCarnegie Mellon UniversityPittsburghUSA

Personalised recommendations