iccv iccv2013 iccv2013-314 iccv2013-314-reference knowledge-graph by maker-knowledge-mining
Source: pdf
Author: Zhuwen Li, Jiaming Guo, Loong-Fah Cheong, Steven Zhiying Zhou
Abstract: This paper addresses real-world challenges in the motion segmentation problem, including perspective effects, missing data, and unknown number of motions. It first formulates the 3-D motion segmentation from two perspective views as a subspace clustering problem, utilizing the epipolar constraint of an image pair. It then combines the point correspondence information across multiple image frames via a collaborative clustering step, in which tight integration is achieved via a mixed norm optimization scheme. For model selection, wepropose an over-segment and merge approach, where the merging step is based on the property of the ?1-norm ofthe mutual sparse representation oftwo oversegmented groups. The resulting algorithm can deal with incomplete trajectories and perspective effects substantially better than state-of-the-art two-frame and multi-frame methods. Experiments on a 62-clip dataset show the significant superiority of the proposed idea in both segmentation accuracy and model selection.
[1] L. Balzano, R. Nowak, and B. Recht. Online identification and tracking of subspaces from highly incomplete information. Technical report, arXiv: 1006.4046, 2011. 7
[2] T. Boult and L. Brown. Factorization-based segmentation of motions. In Proc. of the IEEE Workshop on Motion Understanding, 1991. 1, 2
[3] T. Brox and J. Malik. Object segmentation by long term analysis of point trajectories. In ECCV, 2010. 3
[4] E. J. Cand e`s and T. Tao. The power of convex relaxation: Near-optimal matrix completion. IEEE Trans. Inform. Theory, 56(5):2053–2080, 2009. 2
[5] P. Chen. Optimization algorithms on subspaces: Revisiting missing data problem in low-rank matrix. IJCV, 80(1): 125–142, 2008. 2, 7
[6] T.-J. Chin, H. Wang, and D. Suter. Robust fitting of multiple structures: The statistical learning approach. In ICCV, 2009. 3, 6
[7] J. Costeira and T. Kanade. A multibody factorization method for independently moving objects. IJCV, 29(3):159–179, 1998. 1, 2
[8] R. Dragon, B. Rosenhahn, and J. Ostermann. Multi-scale clustering of frameto-frame correspondences for motion segmentation. In ECCV, 2012. 3, 6
[9] E. Elhamifar and R. Vidal. Sparse subspace clustering. In CVPR, 2009. 1, 2, 4, 6
[10] C. W. Gear. Multibody grouping from motion images. IJCV, 29(2): 133–150, 1998. 1, 2
[11] A. Gruber and Y. Weiss. Multibody factorization with uncertainty and missing data using the em algorithm. In CVPR, 2004. 1, 2
[12] R. Hartley and A. Zisserman. Multiple View Geometry in Computer Vision. Cambridge University Press, 2004. 3
[13] Y. Jian and C. Chen. Two-view motion segmentation with model selection and outlier removal by ransac-enhanced dirichlet process mixture models. IJCV, 88(3):489–2501, 2010. 1, 3
[14] K. Kanatani. Motion segmentation by subspace separation and model selection. In ICCV, 2001. 1, 2
[15] J. Lezama, K. Alahari, J. Sivic, and I. Laptev. Track to the future: Spatiotemporal video segmentation with long-range motions cues. In CVPR, 2011.
[16]
[17]
[18]
[19] 3 H. Li. Two-view motion segmentation from linear programming relaxation. In CVPR, 2007. 1, 3 T. Li, V. Kallem, D. Singaraju, and R. Vidal. Projective factorization ofmultiple rigid-body motions. In CVPR, 2007. 3 Z. Lin, M. Chen, L. Wu, and Y. Ma. The augmented lagrange multiplier method for exact recovery of corrupted low-rank matrices. Technical report, Technical report, UILU-ENG-09-2215, 2009. 2 G. Liu, Z. Lin, S. Yan, J. Sun, Y. Yu, and Y. Ma. Robust recovery of subspace structures by low-rank representation. IEEE TPAMI, (99): 1, 2012. 1, 2, 3, 4, 5, 6
[20] A. Montanari and S. Oh. Matrix completion from a few entries. IEEE Trans. Inform. Theory, 56(6):2988–2998, 2010. 7
[21] A. Rakotomamonjy. Surveying and comparing simultaneous sparse approximation (or group-lasso) algorithms. Signal Processing, 91(7): 1505–1526, 2011. 4
[22] S. Rao, R. Tron, R. Vidal, and Y. Ma. Motion segmentation in the presence of outlying, incomplete, or corrupted trajectories. IEEE TPAMI, 32(10): 1832– 1845, 2010. 1, 2, 6
[23] S. Rao, A. Yang, S. Sastry, and Y. Ma. Robust algebraic segmentation of mixed rigid-body and planar motions from two views. IJCV, 88(3):425–446, 2010. 1, 3
[24] K. Schindler and D. Suter. Two-view multibody structure-and-motion with outliers through model selection. IEEE TPAMI, 28(6):983–995, 2006. 1, 3
[25] K. Schindler, J. U, and H. Wang. Perspective n-view multibody structure-andmotion through model selection. In ECCV, 2006. 3, 6
[26] M. Soltanolkotabi and E. J. Cand e`s. A geometric analysis of subspace clustering with outliers. Annals of Statistics, 40(4):2195–2238, 2011. 2, 3, 5, 6
[27] Y. Sugaya and K. Kanatani. Geometric structure of degeneracy for multi-body motion segmentation. In Workshop on Statistical Methods in Video Processing, 2004. 1, 2
[28] C. Tomasi and T. Kanade. Shape and motion from image streams under orthography. IJCV, 9(2): 137–154, 1992. 1, 2
[29] P. Torr. Bayesian model estimation and selection for epipolar geometry and generic manifold fitting. IJCV, 50(1):35–61, 2002. 1, 3
[30] R. Tron and R. Vidal. A benchmark for the comparison of 3-d motion segmentation algorithms. In CVPR, 2007. 1
[31] R. Vidal and R. Hartley. Motion segmentation with missing data by power factorization and generalized pca. In CVPR, 2004. 1, 2, 6
[32] R. Vidal, Y. Ma, S. Soatto, and S. Sastry. Two-view multibody structure from motion. IJCV, 68(1):7–25, 2006. 1, 3, 6
[33] L. Wolf and A. Shashua. Two-body segmentation from two perspective views. In CVPR, 2001. 1, 3
[34] J. Yan and M. Pollefeys. A general framework for motion segmentation: Independent, articulated, rigid, non-rigid, degenerate and non-degenerate. In ECCV, 2006. 1, 2, 6
[35] T. Zhang, A. Szlam, Y. Wang, and G. Lerman. Hybrid linear modeling via local best-fit flats. IJCV, 100(3):217–240, 2012. 6, 7
[36] D. Zhou and C. Burges. Spectral clustering and transductive learning with multiple views. In ICML, 2007. 4 1376