Publications
• Sorted by Date • Classified by Publication Type • Classified by Research Category •
Matthias Wimmer, Freek Stulp, Sylvia Pietzsch, and Bernd Radig. Learning Local Objective Functions for Robust Face Model Fitting. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 30(8), 2008. To appear.
Download
(unavailable)
Abstract
Model-based fitting has proven to be a successful approach to interpreting thelarge amount of information contained in images. Fitting algorithms search forthe global optimum of an objective function, which should correspond to the bestmodel fit in a given image. Although fitting algorithms have been the subject ofintensive research and evaluation, the objective function is usually designed adhoc, based on implicit and domain-dependent knowledge. This often leads tofunctions with many local minima, and a global minimum that does not correspondto the best model fit.We address the root of this problem by learning more robust objective functions.First, we formulate a set of desirable properties for objective functions, andgive a concrete example of an ideal function that has these properties. Then, wepropose a novel approach that learns an objective function from training datagenerated by this function and manually annotated images. In this approach,critical decisions such as the feature selection are automated, and theremaining manual steps hardly require domain-dependent knowledge. An extensiveempirical evaluation demonstrates that learned objective functions enablefitting algorithms to determine the best model fit more accurately andefficiently than designed objective functions.
BibTeX Entry
@Article{wimmer08learning,
author = {Matthias Wimmer and Freek Stulp and Sylvia Pietzsch and Bernd Radig},
title = {Learning Local Objective Functions for Robust Face Model Fitting},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence~(PAMI)},
year = {2008},
volume = {30},
number = {8},
note = {To appear.},
bib2html_pubtype = {Journal},
bib2html_rescat = {Image Understanding - Face Model Fitting},
abstract = {
Model-based fitting has proven to be a successful approach to interpreting the
large amount of information contained in images. Fitting algorithms search for
the global optimum of an objective function, which should correspond to the best
model fit in a given image. Although fitting algorithms have been the subject of
intensive research and evaluation, the objective function is usually designed ad
hoc, based on implicit and domain-dependent knowledge. This often leads to
functions with many local minima, and a global minimum that does not correspond
to the best model fit.
We address the root of this problem by learning more robust objective functions.
First, we formulate a set of desirable properties for objective functions, and
give a concrete example of an ideal function that has these properties. Then, we
propose a novel approach that learns an objective function from training data
generated by this function and manually annotated images. In this approach,
critical decisions such as the feature selection are automated, and the
remaining manual steps hardly require domain-dependent knowledge. An extensive
empirical evaluation demonstrates that learned objective functions enable
fitting algorithms to determine the best model fit more accurately and
efficiently than designed objective functions.
}
}
Generated by bib2html.pl (written by Patrick Riley ) on Tue Sep 23, 2008 15:09:11 UTC

















