Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Voting strategies for anatomical landmark localization using the implicit shape model

 
: Brauer, Jürgen; Hübner, Wolfgang; Arens, Michael

:
Preprint urn:nbn:de:0011-n-2566287 (2.0 MByte PDF)
MD5 Fingerprint: 9e21efd0cafcfe41c458c7a654729ea1
The original publication is available at springerlink.com
Created on: 5.9.2013


Wilson, R. (Ed.) ; International Association for Pattern Recognition -IAPR-:
Computer Analysis of Images and Patterns. 15th International Conference, CAIP 2013. Vol.1 : York, UK, August 27-29, 2013, Proceedings
Berlin: Springer, 2013 (Lecture Notes in Computer Science 8047)
ISBN: 3-642-40260-7
ISBN: 978-3-642-40260-9 (Print)
ISBN: 978-3-642-40261-6 (Online)
pp.145-153
International Conference on Computer Analysis of Images and Patterns (CAIP) <15, 2013, York>
English
Conference Paper, Electronic Publication
Fraunhofer IOSB ()
human pose estimation; anatomical landmark localization; implicit shape model

Abstract
We address the problem of anatomical landmark localization using monocular camera information only. For person detection the Implicit Shape Model (ISM) is a well known method. Recently it was shown that the same local features that are used to detect persons, can be used to give rough estimates for anatomical landmark locations as well. Though the landmark localization accuracy of the original ISM is far away from being optimal. We show that a direct application of the ISM to the problem of landmark localization leads to poorly localized vote distributions. In this context, we propose three alternative voting strategies which include the use of a reference point, a simple observation vector filtering heuristic, and an observation vector weight learning algorithm. These strategies can be combined in order to further increase localization accuracy. An evaluation on the UMPM benchmark shows that these new voting strategies are able to generate compact and monotonically decreasing vote distributions, which are centered around the ground truth location of the landmarks. As a result, the ratio of correct votes can be increased from only 9.3% for the original ISM up to 42.1% if we combine all voting strategies.

: http://publica.fraunhofer.de/documents/N-256628.html