TOWARDS GENERIC FITTING USING DISCRIMINATIVE ACTIVE APPEARANCE MODELS EMBEDDED ON A RIEMANNIAN MANIFOLD

Pedro Martins, Jorge Batista

2010

Abstract

A solution for Discriminative Active Appearance Models is proposed. The model consists in a set of descriptors which are covariances of multiple features evaluated over the neighborhood of the landmarks whose locations are governed by a Point Distribution Model (PDM). The covariance matrices are a special set of tensors that lie on a Riemannian manifold, which make it possible to measure the dissimilarity and to update them, imposing the temporal appearance consistency. The discriminative fitting method produce patch response maps found by convolution around the current landmark position. Since the minimum of the responce map isn't always the correct solution due to detection ambiguities, our method finds candidates to solutions based on a mean-shift algorithm, followed by an unsupervised clustering technique used to locate and group the candidates. A mahalanobis based metric is used to select the best solution that is consistent with the PDM. Finally the global PDM optimization step is performed using a weighted least-squares warp update, based on the Lucas and Kanade framework. The weights were extracted from a landmark matching score statistics. The effectiveness of the proposed approach was evaluated on unseen data on the challenging Talking Face video sequence, demonstrating the improvement in performance.

References

  1. Baker, S. and Matthews, I. (2004). Lucas-kanade 20 years on: A unifying framework. International Journal of Computer Vision, 56(1):221 - 255.
  2. Baker, S., R.Gross, and I.Matthews (2003). Lucas kanade 20 years on: A unifying framework: Part 3. Technical report, CMU Robotics Institute.
  3. Batur, A. and Hayes, M. (2005). Adaptive active appearance models. IEEE Transactions on Image Processing.
  4. Comaniciu, D. and Meer, P. (2002). Mean shift: A robust approach toward feature space analysis. IEEE PAMI.
  5. D.Cristinacce and T.F.Cootes (2008). Automatic feature localisation with constrained local models. Pattern Recognition.
  6. Figueiredo, M. A. and Jain, A. K. (2002). Unsupervised learning of finite mixture models. IEEE PAMI.
  7. Gross, R., Matthews, I., and Baker, S. (2005). Generic vs. person specific active appearance models. Image and Vision Computing, 23(1):1080-1093.
  8. Matthews, I. and Baker, S. (2004). Active appearance models revisited. IJCV.
  9. Matthews, I., Ishikawa, T., and Baker, S. (2004). The template update problem. IEEE Transactions on Pattern Analysis and Machine Intelligence, 26(1):810 - 815.
  10. Pennec, X., Fillard, P., and Ayache, N. (2006). A riemannian framework for tensor computing. IJCV.
  11. Porikli, F., Tuzel, O., and Meer, P. (2006). Covariance tracking using model update based on lie algebra. In IEEE CVPR.
  12. Sung, J. and Kim, D. (2009). Adaptive active appearance model with incremental learning. Pattern Recognition Letters.
  13. T.F.Cootes, Edwards, G., and C.J.Taylor (2001). Active appearance models. IEEE PAMI.
  14. Wang, Y., Lucey, S., and Cohn, J. (2008a). Enforcing convexity for improved alignment with constrained local models. In IEEE CVPR.
  15. Wang, Y., Lucey, S., Cohn, J., and Saragih, J. M. (2008b). Non-rigid face tracking with local appearance consistency constraint. In IEEE Automatic Face and Gesture Recognition.
Download


Paper Citation


in Harvard Style

Martins P. and Batista J. (2010). TOWARDS GENERIC FITTING USING DISCRIMINATIVE ACTIVE APPEARANCE MODELS EMBEDDED ON A RIEMANNIAN MANIFOLD . In Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2010) ISBN 978-989-674-029-0, pages 363-369. DOI: 10.5220/0002823603630369


in Bibtex Style

@conference{visapp10,
author={Pedro Martins and Jorge Batista},
title={TOWARDS GENERIC FITTING USING DISCRIMINATIVE ACTIVE APPEARANCE MODELS EMBEDDED ON A RIEMANNIAN MANIFOLD},
booktitle={Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2010)},
year={2010},
pages={363-369},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002823603630369},
isbn={978-989-674-029-0},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2010)
TI - TOWARDS GENERIC FITTING USING DISCRIMINATIVE ACTIVE APPEARANCE MODELS EMBEDDED ON A RIEMANNIAN MANIFOLD
SN - 978-989-674-029-0
AU - Martins P.
AU - Batista J.
PY - 2010
SP - 363
EP - 369
DO - 10.5220/0002823603630369