Anisotropic ICP
Anisotropic ICP
Since its introduction in the early 1990s, the Iterative Closest Point (ICP) algorithm has become one of the most well-known methods for geometric alignment of 3D models. Given two roughly aligned shapes represented by two point sets, the algorithm iteratively establishes point correspondences given the current alignment of the data and computes a rigid transformation accordingly. From a statistical point of view, however, it implicitly assumes that the points are observed with isotropic Gaussian noise. [1]
In this research project the first variant of the ICP that accounts for anisotropic localization uncertainty in both input point sets as well as in all steps of the algorithm was presented.
This algorithm was then implemented using MITK. An evaluation on publicly available surface meshes as well as on a set of meshes extracted from medical imaging data shows a dramatic increase in accuracy compared to the original ICP, especially in the case of partial surface registration. [1,2,3]
On this page we present an installer of our implementation of the anisotropic ICP. This installer can be used to test the algorithm. We also provide test data and a user guide.
Downloads
Installer of MITK with Anisotropic ICP Plugin (avaliable for download until 2012, 15th of June):
- Installer (Win32)
- Installer (Win64)
- Installer (Linux)
Test data:
- Test Data Package (zip)
User guide:
- User Guide (pdf)
References
[1] Maier-Hein L, Franz AM, dos Santos TR, Schmidt M, Fangerau M, Meinzer HP, Fitzpatrick JM. Convergent Iterative Closest-Point Algorithm to Accomodate Anisotropic and Inhomogenous Localization Error. In IEEE Trans Pattern Anal Mach Intell. (2011)
[2] Maier-Hein L, dos Santos TR, Franz AM, Meinzer HP, Fitzpatrick JM. Iterative Closest Point Algorithm with Anisotropic Weighting and Its Application to Fine Surface Registration. Proc. SPIE, vol. 7962, pp. 79620. (2011)
[3] Kilgus T, Franz AM, Seitel A, März K, Bartha L, Fangerau M, Mersmann S, Groch A, Meinzer HP, Maier-Hein L. Registration of Partially Overlapping Surfaces for Range Image based Augmented Reality on Mobile Devices. Proc. SPIE, vol. 8316, pp. 831628. (2012)