## The Annals of Applied Statistics

### Statistical analysis of trajectories on Riemannian manifolds: Bird migration, hurricane tracking and video surveillance

#### Abstract

We consider the statistical analysis of trajectories on Riemannian manifolds that are observed under arbitrary temporal evolutions. Past methods rely on cross-sectional analysis, with the given temporal registration, and consequently may lose the mean structure and artificially inflate observed variances. We introduce a quantity that provides both a cost function for temporal registration and a proper distance for comparison of trajectories. This distance is used to define statistical summaries, such as sample means and covariances, of synchronized trajectories and “Gaussian-type” models to capture their variability at discrete times. It is invariant to identical time-warpings (or temporal reparameterizations) of trajectories. This is based on a novel mathematical representation of trajectories, termed transported square-root vector field (TSRVF), and the $\mathbb{L}^{2}$ norm on the space of TSRVFs. We illustrate this framework using three representative manifolds—$\mathbb{S}^{2}$, $\mathrm{SE}(2)$ and shape space of planar contours—involving both simulated and real data. In particular, we demonstrate: (1) improvements in mean structures and significant reductions in cross-sectional variances using real data sets, (2) statistical modeling for capturing variability in aligned trajectories, and (3) evaluating random trajectories under these models. Experimental results concern bird migration, hurricane tracking and video surveillance.

#### Article information

Source
Ann. Appl. Stat., Volume 8, Number 1 (2014), 530-552.

Dates
First available in Project Euclid: 8 April 2014

https://projecteuclid.org/euclid.aoas/1396966297

Digital Object Identifier
doi:10.1214/13-AOAS701

Mathematical Reviews number (MathSciNet)
MR3192001

Zentralblatt MATH identifier
06302246

#### Citation

Su, Jingyong; Kurtek, Sebastian; Klassen, Eric; Srivastava, Anuj. Statistical analysis of trajectories on Riemannian manifolds: Bird migration, hurricane tracking and video surveillance. Ann. Appl. Stat. 8 (2014), no. 1, 530--552. doi:10.1214/13-AOAS701. https://projecteuclid.org/euclid.aoas/1396966297

#### References

• Aggarwal, J. K. and Cai, Q. (1999). Human motion analysis: A review. Comput. Vis. Image Underst. 73 428–440.
• Beg, M. F., Miller, M. I., Trouve, A. and Younes, L. (2005). Computing large deformation metric mappings via geodesic flows of diffeomorphisms. Int. J. Comput. Vis. 61 139–157.
• Bertsekas, D. P. (2007). Dynamic Programming and Optimal Control, 3rd ed. Athena Scientific, Belmont, MA.
• Christensen, G. E. and Johnson, H. J. (2001). Consistent image registration. IEEE Trans. Med. Imag. 20 568–582.
• Dryden, I. L. and Mardia, K. V. (1998). Statistical Shape Analysis. Wiley, Chichester.
• Gavrila, D. M. (1999). The visual analysis of human movement: A survey. Comput. Vis. Image Underst. 73 82–98.
• Jupp, P. E. and Kent, J. T. (1987). Fitting smooth paths to spherical data. J. R. Stat. Soc. Ser. C Appl. Stat. 36 34–46.
• Kenobi, K., Dryden, I. L. and Le, H. (2010). Shape curves and geodesic modelling. Biometrika 97 567–584.
• Kneip, A. and Ramsay, J. O. (2008). Combining registration and fitting for functional models. J. Amer. Statist. Assoc. 103 1155–1165.
• Kume, A., Dryden, I. L. and Le, H. (2007). Shape-space smoothing splines for planar landmark data. Biometrika 94 513–528.
• Kurtek, S., Wu, W. and Srivastava, A. (2011). Signal estimation under random time warpings and its applications in nonlinear signal alignments. Adv. Neural Inf. Process. Syst. 24 676–683.
• Le, H. (2003). Unrolling shape curves. J. Lond. Math. Soc. (2) 68 511–526.
• Le, H. and Kume, A. (2000). The Fréchet mean shape and the shape of the means. Adv. in Appl. Probab. 32 101–113.
• Liu, X. and Müller, H.-G. (2004). Functional convex averaging and synchronization for time-warped random curves. J. Amer. Statist. Assoc. 99 687–699.
• Mardia, K. V. and Jupp, P. E. (2000). Directional Statistics, 3rd ed. Wiley, Chichester.
• Michor, P. W. and Mumford, D. (2007). An overview of the Riemannian metrics on spaces of curves using the Hamiltonian approach. Appl. Comput. Harmon. Anal. 23 74–113.
• Owen, J. C. and Moore, F. R. (2008). Swainson’s thrushes in migratory disposition exhibit reduced immune function. Journal of Ethology 26 383–388.
• Robinson, D. (2012). Functional analysis and partial matching in the square root velocity framework. Ph.D. thesis, Florida State Univ., Tallahassee, FL.
• Shah, J. (2008). $H^{0}$-type Riemannian metrics on the space of planar curves. Quart. Appl. Math. 66 123–137.
• Srivastava, A., Wu, W., Kurtek, S., Klassen, E. and Marron, J. S. (2011a). Registration of functional data using Fisher–Rao metric. Preprint. Available at arXiv:1103.3817v2.
• Srivastava, A., Klassen, E., Joshi, S. H. and Jermyn, I. H. (2011b). Shape analysis of elastic curves in Euclidean spaces. IEEE Trans. Pattern Anal. Mach. Intell. 33 1415–1428.
• Sundaramoorthi, G., Mennucci, A., Soatto, S. and Yezzi, A. (2011). A new geometric metric in the space of curves, and applications to tracking deforming objects by prediction and filtering. SIAM J. Imaging Sci. 4 109–145.
• Trouvé, A. and Younes, L. (2000). On a class of diffeomorphic matching problems in one dimension. SIAM J. Control Optim. 39 1112–1135.
• Tucker, J. D., Wu, W. and Srivastava, A. (2013). Generative models for functional data using phase and amplitude separation. Comput. Statist. Data Anal. 61 50–66.
• Veeraraghavan, A., Srivastava, A., Roy-Chowdhury, A. K. and Chellappa, R. (2009). Rate-invariant recognition of humans and their activities. IEEE Trans. Image Process. 18 1326–1339.
• Younes, L., Michor, P. W., Shah, J. and Mumford, D. (2008). A metric on shape space with explicit geodesics. Atti Accad. Naz. Lincei Cl. Sci. Fis. Mat. Natur. Rend. Lincei (9) Mat. Appl. 19 25–57.