Abstract
Parametric motion models are commonly used in image sequence analysis for different tasks. A robust estimation framework is usually required to reliably compute the motion model over the estimation support in the presence of outliers, while the choice of the right motion model is also important to properly perform the task. However, dealing with model selection within a robust estimation setting remains an open question. We define two original propositions for robust motion-model selection. The first one is an extension of the Takeuchi information criterion. The second one is a new paradigm built from the Fisher statistic. We also derive an interpretation of the latter as a robust Mallows’ \(C_P\) criterion. Both robust motion-model selection criteria are straightforward to compute. We have conducted a comparative objective evaluation on computer-generated image sequences with ground truth, along with experiments on real videos, for the parametric estimation of the 2D dominant motion in an image due to the camera motion. They demonstrate the interest and the efficiency of the proposed robust model-selection methods.
Similar content being viewed by others
References
Agostinelli, C.: Robust model selection in regression via weighted likelihood methodology. Stat. Probab. Lett. 56, 289–300 (2002)
Akaike, H.: A new look at the statistical model identification. IEEE Trans. Autom. Control 19(6), 716–723 (1974)
Ayer, S., Sawhney, H.S.: Layered representation of motion video using robust maximum likelihood estimation of mixture models and MDL encoding. In: ICCV, Cambridge (1995)
Basset, A., Bouthemy, P., Kervrann, C.: Recovery of motion patterns and dominant paths in videos of crowded scenes. In: ICIP, Paris (2014)
Black, M.J., Anandan, P.: The robust estimation of multiple motions: parametric and piecewise-smooth flow fields. Comput. Vis. Image Underst. 63(1), 75–104 (1996)
Bouthemy, P., Toledo Acosta, B.M., Delyon, B.: Robust selection of parametric motion models in image sequences. In: ICIP, Phoenix (2016)
Burnham, K.P., Anderson, D.R.: Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach, 2nd edn. Springer, Berlin (2002)
Cremers, D., Soatto, S.: Motion competition: a variational approach to piecewise parametric motion segmentation. Int. J. Comput. Vis. 62(3), 249–265 (2005)
Fauvet, B., Bouthemy, P., Gros, P., Spindler, F.: A geometrical key-frame selection method exploiting dominant motion estimation in video. In: CVIR, Dublin (2004)
Fortun, D., Bouthemy, P., Kervrann, C.: Optic flow modeling and computation: a survey. Comput. Vis. Image Underst. 134, 1–21 (2015)
François, E., Bouthemy, P.: Derivation of qualitative information in motion analysis. Image Vis. Comput. J. 8(4), 279–287 (1990)
Gheissari, N., Bab-Hadiashar, A.: Motion analysis: model selection and motion segmentation. In: ICIAP, Mantova (2003)
Gheissari, N., Bab-Hadiashar, A.: A comparative study of model selection criteria for computer vision applications. Image Vis. Comput. 26(12), 1636–1649 (2008)
Holland, P.W., Welsch, R.E.: Robust regression using iteratively reweighted least squares. Commun. Stat. Theory Methods 6(9), 813–827 (1977)
Huber, P.J.: Robust Statistics. Wiley Series in Probability and Statistics. Wiley, Hoboken (1981)
Jain, M., van Gemert, J., Jégou, H., Bouthemy, P., Snoek, C.G.M.: Tubelets: unsupervised action proposals from spatiotemporal super-voxels. Int. J. Comput. Vis. 124(3), 287–311 (2017)
Kanatani, K.: Geometric information criterion for model selection. Int. J. Comput. Vis. 26(3), 171–189 (1998)
Kanatani, K.: Model selection for geometric inference. In: ACCV, Melbourne (2002)
Konishi, S., Kitagawa, G.: Information Criteria and Statistical Modeling. Springer Series in Statistics. Springer, Berlin (2008)
Kuss, M.: Gaussian process models for robust regression, classification, and reinforcement learning. Ph.D. Thesis manuscript, TU Darmstadt, March 2006
Machado, J.A.F.: Robust model selection and M-estimation. Econom. Theory 9, 478–493 (1993)
Magri, L., Fusiello, A.: T-linkage: a continuous relaxation of J-linkage for multi-model fitting. In: CVPR, Columbus (2014)
Mallows, C.: Some comments on Cp. Technometrics 15, 661–675 (1973)
Marsland, S., Twining, C.J., Taylor, C.J.: A minimum description length objective function for groupwise non-rigid image registration. Image Vis. Comput. 26(3), 333–346 (2008)
Matsushita, Y., Ofek, E., Ge, W., Tang, X., Shum, H.-Y.: Full-frame video stabilization with motion inpainting. IEEE Trans. Pattern Anal. Mach. Intell. 28(7), 1150–1163 (2006)
Meer, P.: Robust techniques for computer vision. In: Medioni, G., Kang, S.B. (eds.) Emerging Topics in Computer Vision, pp. 107–190. Prentice Hall, Upper Saddle River (2004)
Meyer, F., Bouthemy, P.: Estimation of time-to-collision maps from first order motion models and normal flows. In: ICPR, The Hague (1992)
Müller, S., Welsh, A.H.: Outlier robust model selection in linear regression. J. Am. Stat. Assoc. 100(472), 1297–1310 (2005)
Odobez, J.-M., Bouthemy, P.: Robust multiresolution estimation of parametric motion models. J. Vis. Commun. Image Represent. 6(4), 348–369 (1995)
Perez Rua, J.M., Basset, A., Bouthemy, P.: Detection and localization of anomalous motion in video sequences from local histograms of labeled affine flows. Front. ICT Comput. Image Anal. (2017). https://doi.org/10.3389/fict.2017.00010
Qian, G., Künsch, H.R.: On model selection via stochastic complexity in robust linear regression. J. Stat. Plan. Inference 75, 91–116 (1998)
Rissanen, J.: Modeling by shortest data description. Automatica 14, 465–471 (1978)
Ronchetti, E.: Robust model selection in regression. Stat. Probab. Lett. 3, 21–23 (1985)
Ronchetti, E., Staudte, R.G.: A robust version of Mallows’s Cp. J. Am. Stat. Assoc. 89(426), 550–559 (1994)
Roussas, G.G.: A Course in Mathematical Statistics. Academic Press, Cambrdige (1997)
Saleh, S.: Robust AIC with high breakdown scale estimate. J. Appl. Math., Article ID 286414 (2014)
Schindler, K., Suter, D., Wang, H.: A model-selection framework for multibody structure-and-motion of image sequences. Int. J. Comput. Vis. 79, 159–177 (2008)
Schwarz, G.: Estimating the dimension of a model. Ann. Stat. 6(2), 461–464 (1978)
Senst, T., Eiselen, V., Sikora, T.: Robust local optical flow for feature tracking. IEEE Trans. Circuits Syst. Video Technol. 22(9), 1377–1387 (2012)
Späth, H.: Mathematical Algorithms for Linear Regression. Academic Press, Cambridge (1992)
Szeliski, R.: Image alignment and stitching: a tutorial. Found. Trends Comput. Graph. Comput. Vis. 2(1), 1–104 (2006)
Tennakoon, R.B., Bab-Hadiashar, A., Cao, Z., Hoseinnezhad, R., Suter, D.: Robust model fitting using higher than minimal subset sampling. IEEE Trans. Pattern Anal. Mach. Intell. 38(2), 350–362 (2016)
Tepper, M., Sapiro, G.: Nonnegative matrix underapproximation for robust multiple model fitting. In: CVPR, Honolulu (2017)
Torr, P.: An assessment of information criteria for motion model selection. In: CVPR, San Juan (1997)
Tran, Q.-H., Chin, T.-J., Carneiro, G., Brown, M.S., Suter, D.: In defence of RANSAC for outlier rejection in deformable registration. In: ECCV, Firenze (2012)
Tu, S., Xu, L.: A theoretical investigation of several model selection criteria for dimensionality reduction. Pattern Recognit. Lett. 33(9), 1117–1126 (2012)
Veit, T., Cao, F., Bouthemy, P.: An a contrario decision framework for region-based motion detection. Int. J. Comput. Vis. 68(2), 163–178 (2006)
Waxman, A.M., Kamgar-Parsi, B., Subbarao, M.: Closed-form solutions to image flow equations for 3D structure and motion. Int. J. Comput. Vis. 1, 239–258 (1987)
Wechsler, H., Duric, Z., Li, F., Cherkassky, V.: Motion estimation using statistical learning theory. IEEE Trans. Pattern Anal. Mach. Intell. 26(4), 466–478 (2004)
Yang, J., Li, H.: Dense, accurate optical flow estimation with piecewise parametric model. In: CVPR, Boston (2015)
Yilmaz, A., Javed, O., Shah, M.: Object tracking: a survey. ACM Comput. Surv. 38(4), 13 (2006)
Zaragoza, J., Chin, T.-J., Brown, M.S., Suter, D.: As-projective-as-possible image stitching with moving DLT. In: CVPR, Columbus (2013)
Acknowledgements
This work was supported in part by Conacyt grant for Bertha Mayela Toledo Acosta’s thesis. The authors thank Fabien Spindler for the video acquisitions with the robotic setup.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Bouthemy, P., Toledo Acosta, B.M. & Delyon, B. Robust Model Selection in 2D Parametric Motion Estimation. J Math Imaging Vis 61, 1022–1036 (2019). https://doi.org/10.1007/s10851-019-00883-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10851-019-00883-2