Matches in SemOpenAlex for { <https://semopenalex.org/work/W2007091965> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W2007091965 abstract "In this work, we present a new type of model for object localization, which is well suited for anatomical objectsexhibiting large variability in size, shape and posture, for usage in the discriminative generalized Hough transform(DGHT). The DGHT combines the generalized Hough transform (GHT) with a discriminative training approachto automatically obtain robust and efficient models. It has been shown to be a strong tool for object localizationcapable of handling a rather large amount of shape variability. For some tasks, however, the variability exhibitedby different occurrences of the target object becomes too large to be represented by a standard DGHT model. Tobe able to capture such highly variable objects, several sub-models, representing the modes of variability as seen bythe DGHT, are created automatically and are arranged in a higher dimensional model. The modes of variabilityare identified on-the-fly during training in an unsupervised manner. Following the concept of the DGHT, thesub-models are jointly trained with respect to a minimal localization error employing the discriminative trainingapproach. The procedure is tested on a dataset of thorax radiographs with the target to localize the clavicles.Due to different arm positions, the posture and arrangement of the target and surrounding bones differs strongly,which hampers the training of a good localization model. Employing the new model approach the localizationrate improves by 13% on unseen test data compared to the standard model." @default.
- W2007091965 created "2016-06-24" @default.
- W2007091965 creator A5001054580 @default.
- W2007091965 creator A5004787176 @default.
- W2007091965 creator A5008865065 @default.
- W2007091965 creator A5021686847 @default.
- W2007091965 creator A5023425451 @default.
- W2007091965 creator A5046216922 @default.
- W2007091965 creator A5047068402 @default.
- W2007091965 creator A5080325906 @default.
- W2007091965 date "2012-02-23" @default.
- W2007091965 modified "2023-09-23" @default.
- W2007091965 title "A multi-dimensional model for localization of highly variable objects" @default.
- W2007091965 doi "https://doi.org/10.1117/12.910900" @default.
- W2007091965 hasPublicationYear "2012" @default.
- W2007091965 type Work @default.
- W2007091965 sameAs 2007091965 @default.
- W2007091965 citedByCount "2" @default.
- W2007091965 countsByYear W20070919652013 @default.
- W2007091965 crossrefType "proceedings-article" @default.
- W2007091965 hasAuthorship W2007091965A5001054580 @default.
- W2007091965 hasAuthorship W2007091965A5004787176 @default.
- W2007091965 hasAuthorship W2007091965A5008865065 @default.
- W2007091965 hasAuthorship W2007091965A5021686847 @default.
- W2007091965 hasAuthorship W2007091965A5023425451 @default.
- W2007091965 hasAuthorship W2007091965A5046216922 @default.
- W2007091965 hasAuthorship W2007091965A5047068402 @default.
- W2007091965 hasAuthorship W2007091965A5080325906 @default.
- W2007091965 hasConcept C115961682 @default.
- W2007091965 hasConcept C134306372 @default.
- W2007091965 hasConcept C153180895 @default.
- W2007091965 hasConcept C154945302 @default.
- W2007091965 hasConcept C182365436 @default.
- W2007091965 hasConcept C200518788 @default.
- W2007091965 hasConcept C2781238097 @default.
- W2007091965 hasConcept C31972630 @default.
- W2007091965 hasConcept C33923547 @default.
- W2007091965 hasConcept C41008148 @default.
- W2007091965 hasConcept C97931131 @default.
- W2007091965 hasConceptScore W2007091965C115961682 @default.
- W2007091965 hasConceptScore W2007091965C134306372 @default.
- W2007091965 hasConceptScore W2007091965C153180895 @default.
- W2007091965 hasConceptScore W2007091965C154945302 @default.
- W2007091965 hasConceptScore W2007091965C182365436 @default.
- W2007091965 hasConceptScore W2007091965C200518788 @default.
- W2007091965 hasConceptScore W2007091965C2781238097 @default.
- W2007091965 hasConceptScore W2007091965C31972630 @default.
- W2007091965 hasConceptScore W2007091965C33923547 @default.
- W2007091965 hasConceptScore W2007091965C41008148 @default.
- W2007091965 hasConceptScore W2007091965C97931131 @default.
- W2007091965 hasLocation W20070919651 @default.
- W2007091965 hasOpenAccess W2007091965 @default.
- W2007091965 hasPrimaryLocation W20070919651 @default.
- W2007091965 hasRelatedWork W1480300129 @default.
- W2007091965 hasRelatedWork W2009621568 @default.
- W2007091965 hasRelatedWork W2012796025 @default.
- W2007091965 hasRelatedWork W2036028702 @default.
- W2007091965 hasRelatedWork W2044189432 @default.
- W2007091965 hasRelatedWork W2148821982 @default.
- W2007091965 hasRelatedWork W2151601098 @default.
- W2007091965 hasRelatedWork W2157326618 @default.
- W2007091965 hasRelatedWork W2168518454 @default.
- W2007091965 hasRelatedWork W2232362209 @default.
- W2007091965 hasRelatedWork W2233337389 @default.
- W2007091965 hasRelatedWork W2259489941 @default.
- W2007091965 hasRelatedWork W240204757 @default.
- W2007091965 hasRelatedWork W2460534907 @default.
- W2007091965 hasRelatedWork W2766859027 @default.
- W2007091965 hasRelatedWork W2809341429 @default.
- W2007091965 hasRelatedWork W2942079888 @default.
- W2007091965 hasRelatedWork W2963349720 @default.
- W2007091965 hasRelatedWork W3130422393 @default.
- W2007091965 hasRelatedWork W2735462968 @default.
- W2007091965 isParatext "false" @default.
- W2007091965 isRetracted "false" @default.
- W2007091965 magId "2007091965" @default.
- W2007091965 workType "article" @default.