Abstract
We propose a variational approach for multi-valued velocity field estimation in transparent sequences. Starting from existing local motion estimators, we show a variational model for integrating in space and time these local estimations to obtain a robust estimation of the multi-valued velocity field. With this approach, we can indeed estimate some multi-valued velocity fields which are not necessarily piecewise constant on a layer: Each layer can evolve according to non-parametric optical flow. We show how our approach outperforms some existing approaches, and we illustrate its capabilities on several challenging synthetic/real sequences.
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
Oppenheim, A.V.: Superpostion in a class of nonlinear systems. In: IEEE International Convention, New York, USA, pp. 171–177 (1964)
Bergen, J., et al.: Computing two motions from three frames. In: ICCV 90, Osaka, Japan, December 1990, pp. 27–32 (1990)
Burt, P., Hingorani, R., Kolczynski, R.: Mechanisms for isolating component patterns in the sequential analysis of multiple motion. In: IEEE Workshop on Visual Motion, Princeton, NJ, October 1991, pp. 187–193 (1991)
Irani, M., Peleg, S.: Motion analysis for image enhancement: resolution, occlusion, and transparency. Journal on Visual Communications and Image Representation 4(4), 324–335 (1993)
Irani, M., Rousso, B., Peleg, S.: Computing occluding and transparent motions. IJCV 12(1), 5–16 (1994)
Shizawa, M., Mase, K.: Simultaneous multiple optical flow estimation. In: ICPR 90, vol. 1, pp. 274–278 (1990)
Shizawa, M., Mase, K.: Principle of superposition: a common computational framework for analysis of multiple motion. In: IEEE Workshop on Visual Motion, pp. 164–172 (1991)
Shizawa, M., Mase, K.: A unified computational theory for motion transparency and motion boundaries based on eigenergy analysis. In: CVPR 91, pp. 289–295 (1991)
Liu, H., et al.: Spatio-temporal filters for transparent motion segmentation. In: ICIP 95, pp. 464–468 (1995)
Darrell, T., Simoncelli, E.: Separation of transparent motion into layers using velocity-tuned mechanisms. In: MIT Media Laboratory Vision and Modeling Group Technical Report. Number 244 (1993)
Mota, C., et al.: Divide-and-Conquer strategies for estimating multiple transparent motions. In: Jähne, B., et al. (eds.) IWCM 2004. LNCS, vol. 3417, pp. 66–78. Springer, Heidelberg (2007)
Mühlich, M., Aach, T.: A Theory of Multiple Orientation Estimation. In: Leonardis, A., Bischof, H., Pinz, A. (eds.) ECCV 2006. LNCS, vol. 3952, pp. 69–82. Springer, Heidelberg (2006)
Stuke, I., et al.: Estimation of multiple motions by block matching. In: 4th ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD 2003), pp. 358–362 (2003)
Stuke, I., et al.: Multiple-motion-estimation by block matching using MRF. ACIS, International Journal of Computer and Information Science (2004)
Auvray, V., Bouthemy, P., Lienard, J.: Motion estimation in x-ray image sequence with bi-distributed transparency. In: ICIP 06, Atlanta, USA (2006)
Fitzpatrick, J.: The existence of geometrical density-image transformations corresponding to object motion. CVGIP 44(2), 155–174 (1988)
Black, M., Anandan, P.: The robust estimation of multiple motions: Parametric and piecewise-smooth flow fields. CVGIP: Image Understanding 63(1), 75–104 (1996)
Jepson, A., Black, M.: Mixture models for optical flow computation. In: CVPR 93, pp. 760–761 (1993)
Ju, S., Black, M., Jepson, A.: Skin and bones: Multi-layer, locally affine, optical flow and regularization with transparency. In: Proceedings of CVPR 96, San Francisco, CA, June 1996, pp. 307–314 (1996)
Black, M., Fleet, D., Yacoob, Y.: Robustly estimating changes in image appearance. Computer Vision and Image Understanding 78, 8–31 (2000)
Weiss, Y., Adelson, E.: A unified mixture framework for motion segmentation: incorporating spatial coherence and estimating the number of models. In: CVPR 96, pp. 321–326 (1996)
Rivera, M., Ocegueda, O., Marroquin, J.L.: Entropy controlled gauss-markov random measure field models for early vision. In: Paragios, N., et al. (eds.) VLSM 2005. LNCS, vol. 3752, pp. 137–148. Springer, Heidelberg (2005)
Ramirez-Manzanares, A., et al.: Multi-valued motion fields estimation for transparent sequences with a variational approach. Technical Report RR-5920, INRIA, (Also, Reporte Técnico CIMAT (CC)I-06-12) (June 2006)
Black, M., Rangarajan, P.: On the unification of line processes, outlier rejection, and robust statistics with applications in early vision. IJCV 19(1), 57–91 (1996)
Ramirez-Manzanares, A., Rivera, M.: Brain nerve boundless estimation by restoring and filtering intra-voxel information in DT-MRI. In: VLSM 03, October 2003, pp. 71–80 (2003)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer Berlin Heidelberg
About this paper
Cite this paper
Ramírez-Manzanares, A., Rivera, M., Kornprobst, P., Lauze, F. (2007). A Variational Approach for Multi-valued Velocity Field Estimation in Transparent Sequences. In: Sgallari, F., Murli, A., Paragios, N. (eds) Scale Space and Variational Methods in Computer Vision. SSVM 2007. Lecture Notes in Computer Science, vol 4485. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-72823-8_20
Download citation
DOI: https://doi.org/10.1007/978-3-540-72823-8_20
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-72822-1
Online ISBN: 978-3-540-72823-8
eBook Packages: Computer ScienceComputer Science (R0)