Abstract
Occlusions generally become apparent when integrated over time because violations of the brightness-constancy constraint of optical flow accumulate in occluded areas. Based on this observation, we propose a variational model for occlusion detection that is formulated as an inverse problem. Our forward model adapts the brightness constraint of optical flow to emphasize occlusions by exploiting their temporal behavior, while spatio-temporal regularizers on the occlusion set make our model robust to noise and modeling errors. In terms of minimization, we approximate the resulting variational problem by a sequence of convex optimizations and develop efficient algorithms to solve them. Our experiments show the benefits of the proposed formulation, both forward model and regularizers, in comparison to the state-of-the-art techniques that detect occlusion as the residual of optical-flow estimation.
Similar content being viewed by others
Notes
As the dependence of \(\varvec{u}_{c-j} = \varvec{\bar{u}}_{c-j}(x - \sum _{l=1}^{j-1} \varvec{u}_{c-l}(x) )\) on \(\varvec{u}_{k}\) for \(c-j<k\) disappears when we fix \(\varvec{u}_{c-j}\) to \( \varvec{u}^t_{c-j}~=~\varvec{\bar{u}}_{c-j}(x - \sum _{l=0}^{j-1} \varvec{u}^t_{c-l}(x) )\), we use the standard linearization of optical flow.
References
Adato, Y., Vasilyev, Y., Ben-Shahar, O., Zickler, T.: Toward a theory of shape from specular flow. In: International Conference on Computer Vision (ICCV), pp. 1–8 (2007). doi:10.1109/ICCV.2007.4408883
Alvarez, L., Deriche, R., Papadopoulo, T., Sánchez, J.: Symmetrical dense optical flow estimation with occlusions detection. Int. J. Comput. Vis. 75(3), 371–385 (2007). doi:10.1007/s11263-007-0041-4
Ayvaci, A., Raptis, M., Soatto, S.: Sparse occlusion detection with optical flow. Int. J. Comput. Vis. 97(3), 322–338 (2011). doi:10.1007/s11263-011-0490-7
Ayvaci, A., Soatto, S.: Motion segmentation with occlusions on the superpixel graph. In: IEEE International Conference on Computer Vision, pp. 727–734 (2009). doi:10.1109/ICCVW.2009.5457630
Ayvaci, A., Soatto, S.: Detachable object detection with efficient model selection. In: Energy Minimization Methods in Computer Vision and Pattern Recognition, pp. 191–204 (2011)
Ballester, C., Garrido, L., Lazcano, V., Caselles, V.: A TV-L1 Optical Flow Method with Occlusion Detection. Pattern Recognit. Lecture Notes in Computer Science 7476, 31–40 (2012)
Black, M.J., Fleet, D.J.: Probabilistic detection and tracking of motion boundaries. Int. J. Comput. Vis. 38(3), 231–245 (2000)
Boyd, S., Vandenberghe, L.: Convex Optimization Theory, vol. 25, 7th edn. Cambridge University Press, New York (2004). doi:10.1080/10556781003625177
Brox, T., Bruhn, A., Papenberg, N., Weickert, J.: High accuracy optical flow estimation based on a theory for warping. Eur. Conf. Comput. Vis. 4(May), 25–36 (2004)
Brox, T., Malik, J.: Large displacement optical flow: descriptor matching in variational motion estimation. IEEE Trans. Pattern Anal. Mach. Intell. 33(3), 500–513 (2011). doi:10.1109/TPAMI.2010.143
Butler, D.J., Wulff, J., Stanley, G.B., Black, M.J.: A naturalistic open source movie for optical flow evaluation. In: A. Fitzgibbon, S. Lazebnik, P. Perona, Y. Sato, C. Schmid (eds.) Computer Vision ECCV 2012 SE - 44. Lecture Notes in Computer Science, vol. 7577, pp. 611–625. Springer, Heidelberg (2012). doi:10.1007/978-3-642-33783-3_44
Chambolle, A., Pock, T.: A first-order primal-dual algorithm for convex problems with applications to imaging. J. Math. Imaging Vis. 40(1), 120–145 (2010). doi:10.1007/s10851-010-0251-1
Chan, T.F., Esedoglu, S., Nikolova, M.: Algorithms for finding global minimizers of image segmentation and denoising models. SIAM J. Appl. Math. 66(5), 1632–1648 (2006). doi:10.1137/040615286
Chin, T., Karl, W., Willsky, A.S.: Probabilistic and sequential computation of optical flow using temporal coherence. IEEE Trans. Image Process. 3(6), 773–788 (1994)
Deriche, R., Kornprobst, P., Aubert, G.: Optical-flow estimation while preserving its discontinuities: a variational approach. Recent Dev. Comput. Vis. 9, 69–80 (1996). doi:10.1007/BF01246637
Garg, R., Pizarro, L., Rueckert, D., Agapito, L.: Dense multi-frame optic flow for non-rigid objects using subspace constraints. In: Asian Conference on Computer Vision, pp. 460–473 (2011)
He, X., Yuille, A.: Occlusion boundary detection using pseudo-depth. Comput. Vis. 2010, 539–552 (2010)
Horn, B.K.P.: Robot Vis. MIT Press, Cambridge (1986)
Horn, B.K.P., Schunck, B.G.: Determining optical flow. Artif. Intell. 17(1–3), 185–203 (1981). doi:10.1016/0004-3702(81)90024-2
Humayun, A., Mac Aodha, O., Brostow, G.J.: Learning to find occlusion regions. In: International Conference on Computer Vision and Pattern Recognition, pp. 2161–2168 (2011). doi:10.1109/CVPR.2011.5995517
Ince, S., Konrad, J.: Occlusion-aware optical flow estimation. IEEE Trans. Image Process. 17(8), 1443–1451 (2008). doi:10.1109/TIP.2008.925381
Jackson, J.D., Yezzi, A.J., Soatto, S.: Dynamic shape and appearance modeling via moving and deforming layers. Int. J. Comput. Vis. 79(1), 71–84 (2007). doi:10.1007/s11263-007-0097-1
Kolmogorov, V., Zabih, R.: Computing visual correspondence with occlusions using graph cuts. IEEE Int. Conf. Comput. Vis. 2, 508–515 (2001). doi:10.1109/ICCV.2001.937668
Leordeanu, M., Zanfir, A., Sminchisescu, C.: Locally affine sparse-to-dense matching for motion and occlusion estimation. In: International Conference on Computer Vision (2013)
Lucas, B.D., Kanade, T.: An iterative image registration technique with an application to stereo vision. Imaging Underst. Workshop 130, 121–130 (1981)
Maver, J., Bajcsy, R.: Occlusions as a guide for planning the next view (1993). doi:10.1109/34.211463
Nagel, H.H.: Extending the ’oriented smoothness constraint’ into the temporal domain and the estimation of derivatives of optical flow. In: European Conference on Computer Vision (ECCV), pp. 139–148 (1990)
Pock, T., Unger, M., Cremers, D., Bischof, H.: Fast and exact solution of total variation models on the GPU. Conf. Comput. Vis. Pattern Recognit. Workshop 1, 1–8 (2008). doi:10.1109/CVPRW.2008.4563099
Rockafellar, R.T.: Convex Analysis, vol. 28. Princeton University Press, Princeton (1997)
Roth, S., Domini, F., Black, M.J.: Specular flow and the perception of surface reflectance. Int. Conf. Comput. Vis. Pattern Recognit. 1, 1869–1876 (2006). doi:10.1167/3.9.413
Salgado, A., Sánchez, J.: Optical Flow Estimation with Large Displacements : A Temporal Regularizer. Tech. Rep. April, Cuadernos del Instituto Universitario de Ciencias y Tecnologías Cibernéticas, Las Palmas, España (2006)
Sargin, M.E., Bertelli, L., Manjunath, B.S., Rose, K.: Probabilistic occlusion boundary detection on spatio-temporal lattices. In: International Conference on Computer Vision, pp. 560–567 (2009). doi:10.1109/ICCV.2009.5459190
Smith, P., Drummond, T., Cipolla, R.: Layered motion segmentation and depth ordering by tracking edges. IEEE Trans. Pattern Anal. Mach. Intell. 26(4), 479–494 (2004). doi:10.1109/TPAMI.2004.1265863
Stein, A.N., Hebert, M.: Occlusion boundaries from motion: low-level detection and mid-level reasoning. Int. J. Comput. Vis. 82(3), 325–357 (2009). doi:10.1007/s11263-008-0203-z
Strecha, C., Fransens, R., Gool, L.V.: A probabilistic approach to large displacement optical flow and occlusion detection. Stat. Methods Video Process. 3247, 71–82 (2004)
Sun, D., Roth, S., Black, M.J.: Secrets of optical flow estimation and their principles. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 2432–2439 (2010). doi:10.1109/CVPR.2010.5539939
Sun, D., Sudderth, E.B., Black, M.J.: Layered segmentation and optical flow estimation over time. In: International Conference on Computer Vision and Pattern Recognition, pp. 1768–1775 (2012). doi:10.1109/CVPR.2012.6247873
Sun, J., Li, Y.: Symmetric stereo matching for occlusion handling. IEEE Conf. Comput. Vis. Pattern Recognit. 2, 399–406 (2005)
Sundberg, P., Brox, T.: Occlusion boundary detection and figure/ground assignment from optical flow. In: International Conference on Computer Vision and Pattern Recognition (2011)
Vasilyev, Y., Zickler, T., Gortler, S., Ben-Shahar, O.: Shape from specular flow: is one flow enough? In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 2561–2568 (2011). doi:10.1109/CVPR.2011.5995662
Volz, S., Bruhn, A., Valgaerts, L., Zimmer, H.: Modeling temporal coherence for optical flow. In: 2011 International Conference on Computer Vision, pp. 1116–1123 (2011). doi:10.1109/ICCV.2011.6126359
Wedel, A., Pock, T., Zach, C., Bischof, H., Cremers, D.: An improved algorithm for TV-L 1 optical flow. Stat. Geom. Approach. Vis. Motion Anal. Springer B, 23–45 (2009)
Weinzaepfel, P., Revaud, J., Harchaoui, Z., Schmid, C.: DeepFlow : large displacement optical flow with deep matching. In: International Conference on Computer Vision (2013)
Xiao, J., Cheng, H., Sawhney, H.: Bilateral filtering-based optical flow estimation with occlusion detection. Eur. Conf. Comput. Vis. 3951, 211–224 (2006)
Xiao, J., Shah, M.: Motion layer extraction in the presence of occlusion using graph cuts. IEEE Trans. Pattern Anal. Mach. Intell. 27(10), 1644–1659 (2005). doi:10.1109/TPAMI.2005.202
Xu, L., Jia, J., Matsushita, Y.: Motion detail preserving optical flow estimation. IEEE Trans. Pattern Anal. Mach. Intell. 34(9), 1744–1757 (2012)
Acknowledgments
The authors would like to thank the authors of [2] for sharing their code and D. Davis from UCLA for discussions on optimization. V. Estellers is supported by the Swiss National Science Foundation under Grant P2ELP2_148890.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Estellers, V., Soatto, S. Detecting Occlusions as an Inverse Problem. J Math Imaging Vis 54, 181–198 (2016). https://doi.org/10.1007/s10851-015-0596-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10851-015-0596-6