Accurate Optical Flow Based On Spatiotemporal Gradient Constancy Assumption
Variational methods for optical flow estimation are known for their excellent performance. The method proposed by Brox et al. [5] exemplifies the strength of that framework. It combines several concepts into single energy functional that is then minimized according to clear numerical procedure. In t...
Main Author: | |
---|---|
Format: | Text |
Language: | English |
Published: |
Zenodo
2007
|
Subjects: | |
Online Access: | https://dx.doi.org/10.5281/zenodo.1080677 https://zenodo.org/record/1080677 |
Summary: | Variational methods for optical flow estimation are known for their excellent performance. The method proposed by Brox et al. [5] exemplifies the strength of that framework. It combines several concepts into single energy functional that is then minimized according to clear numerical procedure. In this paper we propose a modification of that algorithm starting from the spatiotemporal gradient constancy assumption. The numerical scheme allows to establish the connection between our model and the CLG(H) method introduced in [18]. Experimental evaluation carried out on synthetic sequences shows the significant superiority of the spatial variant of the proposed method. The comparison between methods for the realworld sequence is also enclosed. : {"references": ["L. Alvarez, J. Weickert, and J. S'anchez. Reliable estimation of dense\noptical flow fields with large displacements. International Journal of Computer Vision, 39(1):41-56, August 2000.", "T. Amiaz and N. Kiryati. Piecewise-smooth dense optical flow via level\nsets. International Journal of Computer Vision, 68(2):111-124, June\n2006.", "G. Aubert, R. Deriche, and P. Kornprobst. Computing optical flow\nvia variational techniques. SIAM Journal on Applied Mathematics,\n60(1):156-182, 1999.", "M. J. Black and P. Anandan. The robust estimation of multiple motions:\nParametric and piecewise-smooth flow fields. Computer Vision and\nImage Understanding, 63(1):75-104, January 1996.", "T. Brox, A. Bruhn, N. Papenberg, and J. Weickert. High accuracy\noptical flow estimation based on a theory for warping. In T. Pajdla\nand J. Matas, editors, Proceedings of the 8th European Conference on\nComputer Vision, volume 3024 of Lecture Notes in Computer Science,\npages 25-36, Prague, Czech Republic, 2004. Springer.", "A. Bruhn and J. Weickert. Towards ultimate motion estimation: Combining\nhighest accuracy with real-time performance. In Proc. Tenth\nInternational Conference on Computer Vision, volume 1, pages 749-\n755, Beijing, China, June 2005. IEEE Computer Society.", "A. Bruhn, J. Weickert, T. Kohlberger, and C. Schn\u252c\u00bforr. A multigrid\nplatform for real-time motion computation with discontinuitypreserving\nvariational methods. International Journal of Computer\nVision, 70(3):257-277, December 2006.", "A. Bruhn, J. Weickert, and C. Schn\u252c\u00bforr. Lucas/Kanade meets\nHorn/Schunck: Combining local and global optic flow methods. International\nJournal of Computer Vision, 61(3):211-231, February 2005.", "I. Cohen. Nonlinear variational method for optical flow computation.\nIn Proc. Eighth Scandinavian Conference on Image Analysis, volume 1,\npages 523-530, Tromso, Norway, May 1993.\n[10] W. Hinterberger, O. Scherzer, C. Schn\u252c\u00bforr, and J. Weickert. Analysis\nof optical flow models in the framework of the calculus of variations.\nNumerical Functional Analysis and Optimization, 23(1/2):69-89, May\n2002.\n[11] B. Horn and B. Schunck. Determining optical flow. Artificial Intelligence,\n17:185-203, August 1981.\n[12] B. D. Lucas and T. Kanade. An iterative image registration technique\nwith an application to stereo vision. In Proc. Seventh International\nJoint Conference on Artificial Intelligence, pages 674-679, Vancouver,\nCanada, August 1981.\n[13] E. M'emin and P. P'erez. A multigrid approach for hierarchical motion\nestimation. In S. Chandran and U. Desai, editors, Proc. Sixth International\nConference on Computer Vision, pages 933-938, Bombay, India,\nJanuary 1998. Narosa Publishing House.\n[14] E. M'emin and P. P'erez. Hierarchical estimation and segmentation\nof dense motion fields. International Journal of Computer Vision,\n46(2):129-155, 2002.\n[15] H.-H. Nagel. Extending the -oriented smoothness constraint- into the\ntemporal domain and the estimation of derivatives of optical flow. In\nO. D. Faugeras, editor, ECCV-90, volume 427 of Lecture Notes in\nComputer Science, pages 139-148. Springer, 1990.\n[16] P. Nesi. Variational approach to optical-flow estimation managing\ndiscontinuities. Image and Vision Computing, 11(7):419-439, September\n1993.\n[17] M. Otte and H.-H. Nagel. Estimation of optical flow based on higherorder\nspatiotemporal derivatives in interlaced and non-interlaced image\nsequences. Artificial Intelligence, 78(1/2):5-43, November 1995.\n[18] A. Rabcewicz. CLG method for optical flow estimation based on\ngradient constancy assumption. In X.-C. Tai, K.-A. Lie, T.F. Chan,\nand S. Osher, editors, Image Processing Based on Partial Differential\nEquations: Proceedings of the International Conference on PDE-Based\nImage Processing and Related Inverse Problems, CMA, Oslo, Norway,\nAugust 8-12, 2005, Mathematics and Visualization. Springer-Verlag,\n2007.\n[19] C. Schn\u252c\u00bforr. On functionals with greyvalue-controlled smoothness terms\nfor determining optical flow. IEEE Transactions on Pattern Analysis\nand Machine Intelligence, 15(10):1074-1079, October 1993.\n[20] S. Uras, F. Girosi, A. Verri, and V. Torre. A computational approach to\nmotion perception. Biological Cybernetics, 60:79-87, 1988.\n[21] J. Weickert and C. Schn\u252c\u00bforr. A theoretical framework for convex\nregularizers in PDE-based computation of image motion. International\nJournal of Computer Vision, 45(3):245-264, December 2001.\n[22] J. Weickert and C. Schn\u252c\u00bforr. Variational optic flow computation with\na spatio-temporal smoothness constraint. Journal of Mathematical\nImaging and Vision, 14(3):245-255, May 2001."]} |
---|