Recovery of Ego-Motion Using Region Alignment

A method for computing the 3D camera motion (the ego-motion) in a static scene is described, where initially a detected 2D motion between two frames is used to align corresponding image regions. We prove that such a 2D registration removes all effects of camera rotation, even for those image regions that remain misaligned. The resulting residual parallax displacement field between the two region-aligned images is an epipolar field centered at the FOE (Focus-of-Expansion). The 3D camera translation is recovered from the epipolar field. The 3D camera rotation is recovered from the computed 3D translation and the detected 2D motion. The decomposition of image motion into a 2D parametric motion and residual epipolar parallax displacements avoids many of the inherent ambiguities and instabilities associated with decomposing the image motion into its rotational and translational components, and hence makes the computation of ego-motion or 3D structure estimation more robust.

[1]  Michael J. Black,et al.  The Robust Estimation of Multiple Motions: Parametric and Piecewise-Smooth Flow Fields , 1996, Comput. Vis. Image Underst..

[2]  P. Anandan,et al.  Parallax Geometry of Pairs of Points for 3D Scene Analysis , 1996, ECCV.

[3]  Nassir Navab,et al.  Relative affine structure: theory and application to 3D reconstruction from perspective views , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[4]  Gilad Adiv,et al.  Determining Three-Dimensional Motion and Structure from Optical Flow Generated by Several Moving Objects , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  D. T. Lawton,et al.  The Use of Difference Fields in Processing Sensor Motion , 1983 .

[6]  P. Anandan,et al.  Direct recovery of shape from multiple views: a parallax based approach , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[7]  Wilfried Enkelmann,et al.  Obstacle detection by evaluation of optical flow fields from image sequences , 1990, Image Vis. Comput..

[8]  Dana H. Ballard,et al.  Computer Vision , 1982 .

[9]  Shmuel Peleg,et al.  Motion Segmentation Using Convergence Properties , 1994 .

[10]  Roberto Cipolla,et al.  Robust structure from motion using motion parallax , 1993, 1993 (4th) International Conference on Computer Vision.

[11]  Chia-Hoang Lee.  Structure And Motion From Two Perspective Views Via Planar Patch , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

[12]  Michal Irani,et al.  Detecting and Tracking Multiple Moving Objects Using Temporal Integration , 1992, ECCV.

[13]  Hans-Hellmut Nagel,et al.  The coupling of rotation and translation in motion estimation of planar surfaces , 1993, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[14]  P. Anandan,et al.  Hierarchical Model-Based Motion Estimation , 1992, ECCV.

[15]  Amnon Shashua,et al.  Projective depth: A geometric invariant for 3D reconstruction from two perspective/orthographic views and for visual recognition , 1993, 1993 (4th) International Conference on Computer Vision.

[16]  K. Hanna.  Direct multi-resolution estimation of ego-motion and structure from motion , 1991, Proceedings of the IEEE Workshop on Visual Motion.

[17]  Shmuel Peleg,et al.  A Three-Frame Algorithm for Estimating Two-Component Image Motion , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[18]  Jan-Olof Eklundh,et al.  Object detection using model based prediction and motion parallax , 1990, ECCV.

[19]  Patrick Bouthemy,et al.  Estimation of time-to-collision maps from first order motion models and normal flows , 1992, [1992] Proceedings. 11th IAPR International Conference on Pattern Recognition.

[20]  Gilad Adiv,et al.  Inherent Ambiguities in Recovering 3-D Motion and Structure from a Noisy Flow Field , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  Harpreet S. Sawhney,et al.  Layered representation of motion video using robust maximum-likelihood estimation of mixture models and MDL encoding , 1995, Proceedings of IEEE International Conference on Computer Vision.

[22]  Michal Irani,et al.  Recovery of ego-motion using image stabilization , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[23]  S. Negahdaripour,et al.  Motion recovery from image sequences using First-order optical flow information , 1991, Proceedings of the IEEE Workshop on Visual Motion.

[24]  H C Longuet-Higgins,et al.  The visual ambiguity of a moving plane , 1984, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[25]  Allan D. Jepson,et al.  Simple method for computing 3D motion and depth , 1990, [1990] Proceedings Third International Conference on Computer Vision.

[26]  Harpreet S. Sawhney,et al.  3D geometry from planar parallax , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[27]  Roberto Cipolla,et al.  Epipole Estimation Using Affine Motion Parallax , 1993, BMVC.