Matrix and Tensor-Based Approximation of 3D Face Animations from Low-Cost Range Sensors
Three-dimensional animation is often represented in the form of a sequence of 3D meshes, also called dynamic animation or Temporally Coherent Mesh Sequence (TCMS). Widespread availability of affordable range sensors makes capturing such data easy, however, its huge volume complicates both storage and further processing. One of the possible solutions is to approximate the data using matrix or tensor decomposition. However the quality the animation may have different impact on both approaches. In this work we use the Microsoft Kinect™ to crate sequences of human face models and compare the approximation error obtained from modelling animations using Principal component analysis (PCA) and Higher Order Singular Value Decomposition (HOSVD). We focus on distortion introduced by reconstruction of data from its truncated factorization. We show that while HOSVD may outperform PCA in terms of approximation error, it may be significantly affected by distortion in animation data.
Keywords3D face models Approximation HOSVD PCA Kinect
This work is partially based on results of the National Science for Research and Development projects: INNOTECH-K2/IN2/50/182645/NCBR/12 and National Science Centre, decision 2011/03/D/ST6/03753. Authors would like to thank Sebastian Opozda for his help with data visualization and development of experimental environment.
- 1.De Aguiar, E., Theobalt, C., Thrun, S., Seidel, H.-P.: Automatic conversion of mesh animations into skeleton-based animations. In: Computer Graphics Forum, vol. 27, no. 2, pp. 389–397. Wiley Online Library (2008)Google Scholar
- 5.Váša, L., Skala, V.: Cobra: compression of the basis for pca represented animations. In: Computer Graphics Forum, vol. 28, no. 6, pp. 1529–1540. Wiley Online Library (2009)Google Scholar
- 6.Breidt, M., Biilthoff, H.H., Curio, C.: Robust semantic analysis by synthesis of 3d facial motion. In: 2011 IEEE International Conference on Automatic Face and Gesture Recognition and Workshops (FG 2011), pp. 713–719. IEEE (2011)Google Scholar
- 8.Romaszewski, M., Głomb, P.: Parameter estimation for hosvd-based approximation of temporally coherent mesh sequences. In: Proceedings of the 11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, pp. 138–145 (2016)Google Scholar
- 9.Romaszewski, M., Gawron, P., Opozda, S.: Dimensionality reduction of dynamic animations using HO-SVD. In: Rutkowski, L., Korytkowski, M., Scherer, R., Tadeusiewicz, R., Zadeh, L.A., Zurada, J.M. (eds.) ICAISC 2014. LNCS (LNAI), vol. 8467, pp. 757–768. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-07173-2_65CrossRefGoogle Scholar
- 11.Jolliffe, I.: Principal Component Analysis. Wiley Online Library (2005)Google Scholar
- 14.Rusinkiewicz, S., Levoy, M.: Efficient variants of the ICP algorithm. In: Proceedings of the Third International Conference on 3-D Digital Imaging and Modeling 2001, pp. 145–152. IEEE (2001)Google Scholar
- 15.Shakhnarovich, G., Darrell, T., Indyk, P.: Nearest-Neighbor Methods in Learning and Vision: Theory and Practice (Neural Information Processing). The MIT press (2006)Google Scholar