Publications
• Sorted by Date • Classified by Publication Type • Classified by Research Category •
Matthias Wimmer, Freek Stulp, Stephan Tschechne, and Bernd Radig. Learning Robust Objective Functions for Model Fitting in Image Understanding Applications. In Proceedings of the British Machine Vision Conference (BMVC 2006), pp. 1159 – 1168, BMVA, September 2006.
Download
Abstract
Model-based methods in computer vision have proven to be a good approach forcompressing the large amount of information in images. Fitting algorithms searchfor those parameters of the model that optimise the objective function given acertain image. Although fitting algorithms have been the subject of intensiveresearch and evaluation, the objective function is usually designed ad hoc andheuristically with much implicit domain-dependent knowledge. This paperformulates a set of requirements that robust objective functions should satisfy.Furthermore, we propose a novel approach that learns the objective function fromtraining images that have been annotated with the preferred model parameters.The requirements are automatically enforced during the learning phase, whichyields generally applicable objective functions. We compare the performance ofour approach to other approaches. For this purpose, we propose a set ofindicators that evaluate how well an objective function meets the statedrequirements.
BibTeX Entry
@InProceedings{wimmer06learning,
author = {Matthias Wimmer and Freek Stulp and Stephan Tschechne and Bernd Radig},
title = {Learning Robust Objective Functions for Model Fitting in Image Understanding Applications},
booktitle = {Proceedings of the British Machine Vision Conference (BMVC 2006)},
editor = {Michael J. Chantler and Emanuel Trucco and Robert B. Fisher},
year = {2006},
month = {September},
publisher = {BMVA},
volume = {3},
pages = {1159 -- 1168},
bib2html_pubtype = {Refereed Conference Paper},
bib2html_rescat = {Image Understanding - Face Model Fitting},
abstract = {
Model-based methods in computer vision have proven to be a good approach for
compressing the large amount of information in images. Fitting algorithms search
for those parameters of the model that optimise the objective function given a
certain image. Although fitting algorithms have been the subject of intensive
research and evaluation, the objective function is usually designed ad hoc and
heuristically with much implicit domain-dependent knowledge. This paper
formulates a set of requirements that robust objective functions should satisfy.
Furthermore, we propose a novel approach that learns the objective function from
training images that have been annotated with the preferred model parameters.
The requirements are automatically enforced during the learning phase, which
yields generally applicable objective functions. We compare the performance of
our approach to other approaches. For this purpose, we propose a set of
indicators that evaluate how well an objective function meets the stated
requirements.
}
}
Generated by bib2html.pl (written by Patrick Riley ) on Tue Sep 23, 2008 15:09:11 UTC

















