Optimal segmentation of dynamic scenes from two perspective views
- 21 November 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2 (10636919) , 281
- https://doi.org/10.1109/cvpr.2003.1211481
Abstract
We present a novel algorithm for optimally segmenting dynamic scenes containing multiple rigidly moving objects. We cast the motion segmentation problem as a constrained nonlinear least squares problem, which minimizes the reprojection error subject to all multibody epipolar constraints. By converting this constrained problem into an unconstrained one, we obtain an objective function that depends on the motion parameters only (fundamental matrices), but is independent on the segmentation of the image features. Therefore, our algorithm does not iterate between feature segmentation and single body motion estimation. Instead, it uses standard nonlinear optimization techniques to simultaneously recover all the fundamental matrices, without prior segmentation. We test our approach on a real sequence.Keywords
This publication has 14 references indexed in Scilit:
- Generalized principal component analysis (GPCA)Published by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Segmentation of dynamic scenes from image intensitiesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Multiple motion scene reconstruction from uncalibrated viewsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Affine 3-D reconstruction from two projective images of independently translating planesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Optimal motion estimation from multiview normalized epipolar constraintPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Reconstruction of a scene with multiple linearly moving objectsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Motion segmentation by subspace separation and model selectionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2001
- Geometric motion segmentation and model selectionPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 1998
- Recovering 3D Shape and Motion from Image Streams Using Nonlinear Least SquaresJournal of Visual Communication and Image Representation, 1994
- Optimal motion and structure estimationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1993