Inlier Estimation for Moving Camera Motion Segmentation
In moving camera videos, motion segmentation is often performed on the optical flow. However, there exist two challenges: (1) Camera motions lead to three primary flows in optical flow: translation, rotation, and radial flow. They are not all solved in existing frameworks under Cartesian coordinate system; (2) A moving camera introduces 3D motion, the depth discontinuities cause the motion discontinuities that severely confuse the motion segmentation. Meanwhile, the mixture of the camera motion and moving objects’ motions make indistinctness between foreground and background. In this work, our solution is to find a low order polynomial to model the background flow field due to its coherence. To this end, we first amend the Helmholts-Hodge Decomposition by adding coherence constraints, which can handle translation, rotation, and radial flow fields under a unified framework. Secondly, we introduce an Incoherence Map and a progressive Quad-Tree partition to reject moving objects and motion discontinuities. Finally, the low order polynomial is achieved from the rest flow samples on two potentials in HHD. We present results on more than twenty videos from four benchmarks. Extensive experiments demonstrate a better performance in dealing with challenging scenes with complex backgrounds. Our method improves the segmentation accuracy of state-of-the-art by \(10\%\sim 30\%\).
KeywordsOptical Flow Camera Motion Radial Flow Motion Segmentation Depth Discontinuity
This work is supported by: Japan Society for the Promotion of Science, Scientific Research KAKENHI for Grant-in-Aid for Young Scientists (ID:25730113).
Supplementary material (avi 1,537 KB)
- 3.Qian, C., Bajic, I.V.: Global motion estimation under translation-zoom ambiguity. In: Proceedings of IEEE PacRim, pp. 46–51 (2013)Google Scholar
- 4.Narayana, M., Hanson, A., Learned-Miller, E.: Coherent motion segmentation in moving camera videos using optical flow orientations. In: ICCV (2013)Google Scholar
- 8.Kwak, S., Lim, T., Nam, W., Han, B., Han, J.H.: Generalized background subtraction based on hybrid inference by belief propagation and bayesian filtering. In: ICCV, pp. 2174–2181 (2011)Google Scholar
- 9.Irani, M., Rousso, B., Peleg, S.: Recovery of ego-motion using image stabilization. In: CVPR (1994)Google Scholar
- 10.Helmholtz, H.: On integrals of the hydrodynamical equations, which express vortex-motion. Philos. Mag. J. Sci. 33, 485–512 (1867)Google Scholar
- 13.Tron, R., Vidal, R.: A benchmark for the comparison of 3D motion segmentation algorithms. In: CVPR (2007)Google Scholar
- 14.Tsai, D., Flagg, M., Rehg, J.M.: Motion coherent tracking with multi-label mrf optimization. In: BMVC (2010)Google Scholar
- 17.Dembczynski, K., Jachnik, A., Kotlowski, W., Waegeman, W., Hullermeier, E.: Optimizing the F-measure in multi-label classication: Plug-in rule approach versus structured loss minimization. In: International Conference on Machine Learning (ICML), pp. 1130–1138 (2013)Google Scholar