Abstract
A 3D facial motion tracking approach has been proposed based on the incorporation of online appearance model (OAM) and cylinder head model (CHM) in the framework of particle filtering. 1) For the construction of OAM, multi-measurements are infused to reduce the influence of lighting and person dependence. 2) The global motion acquired from CHM fitting is set as the initialization of OAM fitting, and the fitting result is set as the initialization of CHM in the next frame. 3) Motion filtering is applied with particle filter combined with local optimization and improved resampling. Objective between subjects evaluations show that our approach is fit to track facial motions.
Similar content being viewed by others
References
Han H, Tong M L, Chen Z C, et al. Variable structure multiple model for articulated human motion tracking from monocular video sequences. Sci China Inf Sci, 2012, 55: 1138–1150
Choi J, Dumortier Y, Choi S, et al. Real-time 3D face tracking and modeling from a webcam. In: Proceedings of IEEE Workshop on Applications of Computer Vision, Breckenridge, 2012. 33–40
La C M, Sclaroff S, Athitsos V. Fast, reliable head tracking under varying illumination: an approach based on registration of texture mapped 3D models. IEEE Trans Pattern Anal, 2000, 22: 322–336
Sung J, Kanade T, Kim D. Pose robust face tracking by combining active appearance models and cylinder head models. Int J Comput Vision, 2008, 80: 260–274
Lui Y M, Beveridge J R, Whitley L D. Adaptive appearance model and condensation algorithm for robust face tracking. IEEE Trans Syst Man Cy A, 2010, 40: 437–448
Zhou S K, Chellappa R, Moghaddam B. Visual tracking and recognition using appearance-adaptive models in particle filters. IEEE Trans Image Process, 2004, 13: 1491–1506
Dornaika F, Davoine F. On appearance based face and facial action tracking. IEEE Trans Circ Syst Vid, 2006, 16: 1107–1124
Wen Z, Huang T S. Capturing subtle facial motions in 3D face tracking. In: Proceedings of 9th IEEE International Conference on Computer Vision, Nice, 2003. 1343–1350
Taheri S, Sankaranarayanan A C, Chellappa R. Joint albedo estimation and pose tracking from video. IEEE Trans Pattern Anal, 2012, 54: 1674–1689
Marks T K, Hershey J R, Movellan J R. Tracking motion, deformation, and texture using conditionally Gaussian processes. IEEE Trans Pattern Anal, 2010, 32: 348–363
Xiao J, Moriyama T, Kanade T, et al. Robust full-motion recovery of head by dynamic templates and registration techniques. Int J Imag Syst Tech, 2003, 13: 85–94
Surya T T, Robert E K. Importance sampling: a review. Wiley Computation Stat, 2009, 2: 54–60
Ahlberg J. CANDIDE-3-An Updated Parameterized Face. Technical Report LiTH-ISY-R-2326, 2001
Fidaleo D, Medioni G, Fua P, et al. An investigation of model bias in 3D face tracking. In: Proceedings of International Conference on Analysis and Modeling of Faces and Gestures, Beijing, 2005. 125–139
Grassberger P. Pruned-enriched rosenbluth method: simulations of θ polymers of chain length up to 1000000. Phys Rev E, 1997, 56: 3682–3693
Hu Y K, Wang Z F. A low-dimensional illumination space representation of human faces for arbitrary lighting conditions. Int Conf Pattern Recogn, 2006, 33: 9–14
Nordstrom M M, Larsen M, Sierakowski J, et al. The IMM Face Database — An Annotated Dataset of 240 Face Images. Informatics and Mathematical Modelling, Technical University of Denmark, DTU, 2004
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Yu, J., Wang, Z. 3D facial motion tracking by combining online appearance model and cylinder head model in particle filtering. Sci. China Inf. Sci. 57, 1–7 (2014). https://doi.org/10.1007/s11432-013-5023-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11432-013-5023-2