Object recognition by computer: the role of geometric constraints
Object recognition by computer: the role of geometric constraints
Object pose from 2-D to 3-D point and line correspondences
International Journal of Computer Vision
New algebraic tools for classical geometry
Geometric computing with Clifford algebras
The geometry and algebra of kinematics
Geometric computing with Clifford algebras
The motor extended Kalman filter for dynamic rigid motion estimation from line observations
Geometric computing with Clifford algebras
Three D-Dynamic Scene Analysis: A Stereo Based Approach
Three D-Dynamic Scene Analysis: A Stereo Based Approach
Analysis of Orientation Problems Using Plucker Lines
ICPR '98 Proceedings of the 14th International Conference on Pattern Recognition-Volume 1 - Volume 1
Numerical Methods for Model-Based Pose Recovery
Numerical Methods for Model-Based Pose Recovery
Adaptive Pose Estimation for Different Corresponding Entities
Proceedings of the 24th DAGM Symposium on Pattern Recognition
| Hi-index | 0.00 |
The paper concerns 2D-3D pose estimation in the algebraic language of kinematics. The pose estimation problem is modeled on the base of several geometric constraint equations. In that way the projective geometric aspect of the topic is implicitly represented and thus, pose estimation is a pure kinematic problem. The authors propose the use of motor algebra to model screw displacements of lines or the use of rotor algebra to model the motion of points. Instead of using matrix based LMS optimization, the development of special extended Kalman filters is proposed. In this paper extended Kalman filters for estimating rotation and translation of several constraints in terms of rotors and motors will be presented. The experiments aim to compare the use of different constraints and different methods of optimal estimating the pose parameters.