by M Wimmer, F Stulp, S Tschechne and B Radig
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.
Reference:
Learning Robust Objective Functions for Model Fitting in Image Understanding Applications (M Wimmer, F Stulp, S Tschechne and B Radig), In Proceedings of the 17th British Machine Vision Conference (BMVC) (MJ. Chantler, E Trucco, RB. Fisher, eds.), BMVA, volume 3, 2006.
Bibtex Entry:
@inproceedings{wimmer_learning_2006,
author = {M Wimmer and F Stulp and S Tschechne and B Radig},
title = {Learning Robust Objective Functions for Model Fitting in Image Understanding
Applications},
booktitle = {Proceedings of the 17th British Machine Vision Conference ({BMVC)}},
year = {2006},
editor = {Chantler, Michael J. and Trucco, Emanuel and Fisher, Robert B.},
volume = {3},
pages = {1159--1168},
address = {Edinburgh, {UK}},
month = {sep},
publisher = {{BMVA}},
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.},
}