Abstract
Discriminative approaches for human pose estimation model the functional mapping, or conditional distribution, between image features and 3D pose. Learning such multi-modal models in high dimensional spaces, however, is challenging with limited training data; often resulting in over-fitting and poor generalization. To address these issues latent variable models (LVMs) have been introduced. Shared LVMs attempt to learn a coherent, typically non-linear, latent space shared by image features and 3D poses, distribution of data in that latent space, and conditional distributions to and from this latent space to carry out inference. Discovering the shared manifold structure can, in itself, however, be challenging. In addition, shared LVMs models are most often non-parametric, requiring the model representation to be a function of the training set size. We present a parametric framework that addresses these shortcoming. In particular, we learn latent spaces, and distributions within them, for image features and 3D poses separately first, and then learn a multi-modal conditional density between these two low-dimensional spaces in the form of Gaussian Mixture Regression. Using our model we can address the issue of over-fitting and generalization, since the data is denser in the learned latent space, as well as avoid the necessity of learning a shared manifold for the data. We quantitatively evaluate and compare the performance of the proposed method to several state-of-the-art alternatives, and show that our method gives a competitive performance.
Keywords
- Latent Space
- Gaussian Mixture Model
- Canonical Correlation Analysis
- Latent Variable Model
- Locality Preserve Projection
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.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Sidenbladh, H., Black, M., Fleet, D.: Stochastic tracking of 3D human figures using 2D image motion. In: Vernon, D. (ed.) ECCV 2000. LNCS, vol. 1843, pp. 702–718. Springer, Heidelberg (2000)
Sminchisescu, C., Triggs, B.: Covariance scaled sampling for monocular 3d body tracking. In: CVPR (2001)
Agarwal, A., Triggs, B.: 3d human pose from silhouettes by relevance vector regression. In: CVPR (2004)
Agarwal, A., Triggs, B.: Monocular human motion capture with a mixture of regressors. In: CVPR (2005)
Bissacco, A., Yang, M., Soatto, S.: Fast human pose estimation using appearance and motion via multi-dimensional boosting regression. In: CVPR (2007)
Bo, L., Sminchisescu, C.: Structured output-associative regression. In: CVPR (2009)
Elgammal, A.M., Lee, C.S.: Inferring 3d body pose from silhouettes using activity manifold learning. In: CVPR (2004)
Fathi, A., Mori, G.: Human pose estimation using motion exemplars. In: ICCV (2007)
Guo, F., Qian, G.: Learning and inference of 3d human poses from gaussian mixture modeled silhouettes. In: ICPR (2006)
Jaeggli, T., Koller-Meier, E., Van Gool, L.: Learning Generative Models for Multi-Activity Body Pose Estimation. International Journal of Computer Vision 83, 121–134 (2009)
Kanaujia, A., Sminchisescu, C., Metaxas, D.: Spectral latent variable models for perceptual inference. In: ICCV (2007)
Navaratnam, R., Fitzgibbon, A., Cipolla, R.: The Joint Manifold Model for Semi-supervised Multi-valued Regression. In: ICCV (2007)
Shakhnarovich, G., Viola, P., Darrell, T.: Fast pose estimation with parameter-sensitive hashing. In: ICCV (2003)
Sminchisescu, C., Kanaujia, A., Li, Z., Metaxas, D.: Discriminative density propagation for 3d human motion estimation. In: CVPR (2005)
Sminchisescu, C., Kanaujia, A., Metaxas, D.: Learning joint top-down and bottom-up processes for 3d visual inference. In: CVPR (2006)
Urtasun, R., Darrell, T.: Sparse probabilistic regression for activity-independent human pose inference. In: CVPR (2008)
Ek, C., Torr, P., Lawrence, N.: Gaussian process latent variable models for human pose estimation. In: Popescu-Belis, A., Renals, S., Bourlard, H. (eds.) MLMI 2007. LNCS, vol. 4892, pp. 132–143. Springer, Heidelberg (2008)
Sigal, L., Memisevic, R., Fleet, D.J.: Shared Kernel Information Embedding for Discriminative Inference. In: CVPR (2009)
He, X., Niyogi, P.: Locality preserving projections. In: NIPS (2003)
Nadaraya, E.: On estimation regression. Theory of Probability and its Applications 9, 141–142 (1964)
Sminchisescu, C., Kanaujia, A., Li, Z., Metaxas, D.: Learning to reconstruct 3d human motion from bayesian mixtures of experts: A probabilistic discriminative approach. Technical Report CSRG-502, University of Toronto (2004)
Thorndike, R.: Canonical correlation analysis. Applied Multivariate Statistics and Mathematical Modeling, 237–263 (2000)
Hotelling, H.: Relations between two sets of variates. Biometrika 28, 321–377 (1936)
Tian, T., Li, R., Sclaroff, S.: Articulated pose estimation in a learned smooth space of feasible solutions. In: Worshop on Learning in Computer Vision and Pattern Recognition, San Diego (2005)
E–frontier. Curious Labs Poser. Computer Software
CMU Motion Capture Database, http://mocap.cs.cmu.edu/
Sigal, L., Balan, A., Black, M.: Combined discriminative and generative articulated pose and non-rigid shape estimation. In: NIPS (2007)
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
Tian, Y., Sigal, L., Badino, H., De la Torre, F., Liu, Y. (2011). Latent Gaussian Mixture Regression for Human Pose Estimation. In: Kimmel, R., Klette, R., Sugimoto, A. (eds) Computer Vision – ACCV 2010. ACCV 2010. Lecture Notes in Computer Science, vol 6494. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-19318-7_53
Download citation
DOI: https://doi.org/10.1007/978-3-642-19318-7_53
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-19317-0
Online ISBN: 978-3-642-19318-7
eBook Packages: Computer ScienceComputer Science (R0)