Iterative Estimation of Rigid-Body Transformations Application to Robust Object Tracking and Iterative Closest Point

Closed-form solutions are traditionally used in computer vision for estimating rigid body transformations. Here we suggest an iterative solution for estimating rigid body transformations and prove its global convergence. We show that for a number of applications involving repeated estimations of rigid body transformations, an iterative scheme is preferable to a closed-form solution. We illustrate this experimentally on two applications, 3D object tracking and image registration with Iterative Closest Point. Our results show that for those problems using an iterative and continuous estimation process is more robust than using many independent closed-form estimations.


Published in:
Journal Of Mathematical Imaging And Vision, 43, 1-9
Year:
2012
Publisher:
Springer Verlag
ISSN:
0924-9907
Keywords:
Laboratories:




 Record created 2012-05-04, last modified 2018-03-17


Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)