Mutual information for lucas-kanade tracking (milk): An inverse compositional formulation (2008)
| Venue: | In IEEE Trans. on Pattern Analysis and Machine Intelligence |
| Citations: | 6 - 1 self |
BibTeX
@INPROCEEDINGS{Dowson08mutualinformation,
author = {Nicholas Dowson and Richard Bowden and Senior Member},
title = {Mutual information for lucas-kanade tracking (milk): An inverse compositional formulation},
booktitle = {In IEEE Trans. on Pattern Analysis and Machine Intelligence},
year = {2008}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
Abstract—Mutual Information (MI) is popular for registration via function optimization. This work proposes an inverse compositional formulation of MI for Levenberg-Marquardt optimization. This yields a constant Hessian, which may be precomputed. Speed improvements of 15 percent were obtained, with convergence accuracies similar those of the standard formulation. Index Terms—Mutual Information, registration, Newton optimization, tracking.
