Motion Estimation of Articulated Objects from Perspective Views

Motion estimation of articulated objects with two subparts from monocular images are studied in this paper for three cases: 1) one subpart translates, and the other one rotates around the joint; 2) the two rotation axes of the subparts are parallel to each other; 3) the two rotation axes of the subparts are perpendicular with each other. Three motion models are established respectively, and the conditions for a solution are discussed in detail, which shows that only 4, 5 and 6 image point correspondences are needed respectively for the three kinds of articulated motion estimation. The problem of how to distribute the points on the two subparts is also explained. Finally, a lot of simulated experiments are presented, validating the rightness and efficiency of our motion models.

[1]  H. C. Longuet-Higgins,et al.  A computer algorithm for reconstructing a scene from two projections , 1981, Nature.

[2]  Takashi Totsuka,et al.  Constraint-conscious smoothing framework for the recovery of 3D articulated motion from image sequences , 2000, Proceedings Fourth IEEE International Conference on Automatic Face and Gesture Recognition (Cat. No. PR00580).

[3]  Jitendra Malik,et al.  Tracking people with twists and exponential maps , 1998, Proceedings. 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No.98CB36231).

[4]  Jake K. Aggarwal,et al.  Structure from Motion of Rigid and Jointed Objects , 1981, Artif. Intell..

[5]  Camillo J. Taylor,et al.  Reconstruction of articulated objects from point correspondences in a single uncalibrated image , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[6]  S Chen,et al.  Motion analysis of articulated objects and its applications in human motion analysis , 2000 .

[7]  Thomas S. Huang,et al.  Motion and structure from feature correspondences: a review , 1994, Proc. IEEE.

[8]  David Sinclair,et al.  Further constraints on visual articulated motions , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.