DMCA
Non-rigid point set registration: Coherent Point Drift (CPD) (2006)
Cached
Download Links
- [faculty.ucmerced.edu]
- [arxiv.org]
- [arxiv.org]
- [books.nips.cc]
- [faculty.ucmerced.edu]
- [machinelearning.wustl.edu]
- [graphics.stanford.edu]
- DBLP
Other Repositories/Bibliography
| Venue: | IN ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 19 |
| Citations: | 141 - 0 self |
BibTeX
@INPROCEEDINGS{Myronenko06non-rigidpoint,
author = {Andriy Myronenko and Xubo Song and Á. Carreira-Perpiñán},
title = {Non-rigid point set registration: Coherent Point Drift (CPD)},
booktitle = {IN ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 19},
year = {2006},
publisher = {MIT Press}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We introduce Coherent Point Drift (CPD), a novel probabilistic method for non-rigid registration of point sets. The registration is treated as a Maximum Likelihood (ML) estimation problem with motion coher-ence constraint over the velocity field such that one point set moves coherently to align with the second set. We formulate the motion coherence constraint and derive a solution of regularized ML estimation through the variational approach, which leads to an elegant kernel form. We also derive the EM algorithm for the penalized ML optimization with deterministic annealing. The CPD method simultaneously finds both the non-rigid transformation and the correspondence between two point sets without making any prior assumption of the transformation model except that of motion coherence. This method can estimate complex non-linear non-rigid transformations, and is shown to be accurate on 2D and 3D examples and robust in the presence of outliers and missing points.
Keyphrases
coherent point drift non-rigid point point set motion coherence constraint transformation model estimation problem second set variational approach regularized ml estimation maximum likelihood deterministic annealing elegant kernel form motion coherence motion coher-ence constraint non-rigid transformation cpd method em algorithm velocity field prior assumption novel probabilistic method complex non-linear non-rigid transformation penalized ml optimization non-rigid registration
