Abstract
A novel nonlinear probabilistic feature extraction method, called Spatio-Temporal Gaussian Process Latent Variable Model, is introduced to discover generalised and continuous low dimensional representation of multivariate time series data in the presence of stylistic variations. This is achieved by incorporating a new spatio-temporal constraining prior over latent spaces within the likelihood optimisation of Gaussian Process Latent Variable Models (GPLVM). As a result, the core pattern of multivariate time series is extracted, whereas a style variability is marginalised. We validate the method by qualitative comparison of different GPLVM variants with their proposed spatio-temporal versions. In addition we provide quantitative results on a classification application, i.e. view-invariant action recognition, where imposing spatio-temporal constraints is essential. Performance analysis reveals that our spatio-temporal framework outperforms the state of the art.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Belkin, M., Niyogi, P.: Laplacian eigenmaps and spectral techniques for embedding and clustering. In: Proc. NIPS, vol. 14, pp. 585–591 (2001)
Ek, C., Torr, P., Lawrence, N.D.: Gaussian process latent variable models for human pose estimation. Machine Learning for Multimodal Interaction, 132–143 (2007)
Jenkins, O., Matarić, M.: A spatio-temporal extension to isomap nonlinear dimension reduction. In: Proc. ICML, pp. 441–448 (2004)
Junejo, I., Dexter, E., Laptev, I., Pérez, P.: Cross-view action recognition from temporal self-similarities. In: Proc. ECCV, vol. 12 (2008)
Lawrence, N.: Gaussian process latent variable models for visualisation of high dimensional data. In: Proc. NIPS, vol. 16 (2004)
Lawrence, N.: Learning for larger datasets with the Gaussian process latent variable model. In: Proc. AISTATS (2007)
Lewandowski, J., Makris, D., Nebel, J.C.: View and style-independent action manifolds for human activity recognition. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010. LNCS, vol. 6316, pp. 547–560. Springer, Heidelberg (2010)
Lewandowski, M., Martinez-del-Rincon, J., Makris, D., Nebel, J.-C.: Temporal extension of laplacian eigenmaps for unsupervised dimensionality reduction of time series. In: Proc. ICPR (2010)
Liu, J., Ali, S., Shah, M.: Recognizing human actions using multiple features. In: Proc. CVPR (2008)
Liu, J., Shah, M.: Learning human actions via information maximization. In: Proc. CVPR (2008)
Möller, M.: A scaled conjugate gradient algorithm for fast supervised learning. Neural Networks 6(4), 525–533 (1993)
Lawrence, N.D., Quinonero-Candela, J.: Local Distance Preservation in the GP-LVM Through Back Constraints. In: Proc. ICML, pp. 513–520 (2006)
Roweis, S., Saul, L.: Nonlinear dimensionality reduction by locally linear embedding. Science 290(5500), 2323–2326 (2000)
Schölkopf, B., Smola, A., Müller, K.: Kernel principal component analysis. In: ICANN, pp. 583–588 (1997)
Shyr, A., Urtasun, R., Jordan, M.: Sufficient dimension reduction for visual sequence classification. In: Proc. CVPR (2010)
Sigal, L., Black, M.: HumanEva: Synchronized Video and Motion Capture Dataset for Evaluation of Articulated Human Motion. Brown Univertsity (2006)
Tenenbaum, J., Silva, V., Langford, J.: A global geometric framework for nonlinear dimensionality reduction. Science 290(5500), 2319–2323 (2000)
Tipping, M., Bishop, C.: Probabilistic principal component analysis. Journal of the Royal Statistical Society, Series B 61, 611–622 (1999)
Urtasun, R., Darrell, T.: Discriminative Gaussian process latent variable model for classification. In: Proc. ICML, pp. 927–934 (2007)
Urtasun, R., Fleet, D.J., Fua, P.: 3D people tracking with gaussian process dynamical models. In: Proc. CVPR, vol. 1, pp. 238–245 (2006)
Urtasun, R., Fleet, D., Geiger, A., Popović, J., Darrell, T., Lawrence, N.: Topologically-constrained latent variable models. In: Proc. ICML (2008)
Wang, J., Fleet, D., Hertzmann, A.: Gaussian process dynamical models. In: Proc. NIPS, vol. 18, pp. 1441–1448 (2006)
Weinland, D., Boyer, E., Ronfard, R.: Action recognition from arbitrary views using 3D exemplars. In: Proc. ICCV, vol. 5(7), p. 8 (2007)
Yan, P., Khan, S., Shah, M.: Learning 4D action feature models for arbitrary view action recognition. In: Proc. CVPR, vol. 12 (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lewandowski, M., Makris, D., Nebel, JC. (2011). Probabilistic Feature Extraction from Multivariate Time Series Using Spatio-Temporal Constraints. In: Huang, J.Z., Cao, L., Srivastava, J. (eds) Advances in Knowledge Discovery and Data Mining. PAKDD 2011. Lecture Notes in Computer Science(), vol 6635. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-20847-8_15
Download citation
DOI: https://doi.org/10.1007/978-3-642-20847-8_15
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-20846-1
Online ISBN: 978-3-642-20847-8
eBook Packages: Computer ScienceComputer Science (R0)