Object recognition and localization via pose clustering
Computer Vision, Graphics, and Image Processing
Numerical recipes in C: the art of scientific computing
Numerical recipes in C: the art of scientific computing
Fitting Parameterized Three-Dimensional Models to Images
IEEE Transactions on Pattern Analysis and Machine Intelligence
Monocular pose of a rigid body using point landmarks
CVGIP: Image Understanding
Review and analysis of solutions of the three point perspective pose estimation problem
International Journal of Computer Vision
Finding the parts of objects in range images
Computer Vision and Image Understanding
Surface recovery from range images using curvature and motion consistency
Computer Vision and Image Understanding
3D Pose Estimation by Directly Matching Polyhedral Models to Gray Value Gradients
International Journal of Computer Vision
Darboux Frames, Snakes, and Super-Quadrics: Geometry from the Bottom Up
IEEE Transactions on Pattern Analysis and Machine Intelligence
3-D Pose from 3 Points Using Weak-Perspective
IEEE Transactions on Pattern Analysis and Machine Intelligence
Uniqueness of 3D Pose Under Weak Perspective: A Geometrical Proof
IEEE Transactions on Pattern Analysis and Machine Intelligence
Autonomous recognition: driven by ambiguity
CVPR '96 Proceedings of the 1996 Conference on Computer Vision and Pattern Recognition (CVPR '96)
Active Recognition: Using Uncertainty to Reduce Ambiguity
ICPR '96 Proceedings of the 1996 International Conference on Pattern Recognition (ICPR '96) Volume I - Volume 7270
Uncertainty in Pose Estimation: A Bayesian Approach
ICPR '98 Proceedings of the 14th International Conference on Pattern Recognition-Volume 2 - Volume 2
Hi-index | 0.00 |
The problem of estimating and tracking the pose of a 3-D object is a well-established problem in machine vision with important applications in terrestrial and space robotics. This paper describes how 3-D range data, available form a new generation of real-time laser rangefinding systems, can be used to solve the pose determination problem. The approach is based on analysis of the local geometric structure encoded in the range data to extract landmarks. Local configurations of these landmarks provide estimates of identity and pose through matching against a nominal models using a Bayesian optimization technique. Aggregates of local estimates are used to provide a robust estimate of global pose. The technique is well-suited to space tracking applications for which examples are provided.