Cargando…

Local shape feature fusion for improved matching, pose estimation and 3D object recognition

We provide new insights to the problem of shape feature description and matching, techniques that are often applied within 3D object recognition pipelines. We subject several state of the art features to systematic evaluations based on multiple datasets from different sources in a uniform manner. We...

Descripción completa

Detalles Bibliográficos
Autores principales: Buch, Anders G., Petersen, Henrik G., Krüger, Norbert
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Springer International Publishing 2016
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4783326/
https://www.ncbi.nlm.nih.gov/pubmed/27066334
http://dx.doi.org/10.1186/s40064-016-1906-1
_version_ 1782420085962440704
author Buch, Anders G.
Petersen, Henrik G.
Krüger, Norbert
author_facet Buch, Anders G.
Petersen, Henrik G.
Krüger, Norbert
author_sort Buch, Anders G.
collection PubMed
description We provide new insights to the problem of shape feature description and matching, techniques that are often applied within 3D object recognition pipelines. We subject several state of the art features to systematic evaluations based on multiple datasets from different sources in a uniform manner. We have carefully prepared and performed a neutral test on the datasets for which the descriptors have shown good recognition performance. Our results expose an important fallacy of previous results, namely that the performance of the recognition system does not correlate well with the performance of the descriptor employed by the recognition system. In addition to this, we evaluate several aspects of the matching task, including the efficiency of the different features, and the potential in using dimension reduction. To arrive at better generalization properties, we introduce a method for fusing several feature matches with a limited processing overhead. Our fused feature matches provide a significant increase in matching accuracy, which is consistent over all tested datasets. Finally, we benchmark all features in a 3D object recognition setting, providing further evidence of the advantage of fused features, both in terms of accuracy and efficiency.
format Online
Article
Text
id pubmed-4783326
institution National Center for Biotechnology Information
language English
publishDate 2016
publisher Springer International Publishing
record_format MEDLINE/PubMed
spelling pubmed-47833262016-04-09 Local shape feature fusion for improved matching, pose estimation and 3D object recognition Buch, Anders G. Petersen, Henrik G. Krüger, Norbert Springerplus Research We provide new insights to the problem of shape feature description and matching, techniques that are often applied within 3D object recognition pipelines. We subject several state of the art features to systematic evaluations based on multiple datasets from different sources in a uniform manner. We have carefully prepared and performed a neutral test on the datasets for which the descriptors have shown good recognition performance. Our results expose an important fallacy of previous results, namely that the performance of the recognition system does not correlate well with the performance of the descriptor employed by the recognition system. In addition to this, we evaluate several aspects of the matching task, including the efficiency of the different features, and the potential in using dimension reduction. To arrive at better generalization properties, we introduce a method for fusing several feature matches with a limited processing overhead. Our fused feature matches provide a significant increase in matching accuracy, which is consistent over all tested datasets. Finally, we benchmark all features in a 3D object recognition setting, providing further evidence of the advantage of fused features, both in terms of accuracy and efficiency. Springer International Publishing 2016-03-08 /pmc/articles/PMC4783326/ /pubmed/27066334 http://dx.doi.org/10.1186/s40064-016-1906-1 Text en © Buch et al. 2016 Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
spellingShingle Research
Buch, Anders G.
Petersen, Henrik G.
Krüger, Norbert
Local shape feature fusion for improved matching, pose estimation and 3D object recognition
title Local shape feature fusion for improved matching, pose estimation and 3D object recognition
title_full Local shape feature fusion for improved matching, pose estimation and 3D object recognition
title_fullStr Local shape feature fusion for improved matching, pose estimation and 3D object recognition
title_full_unstemmed Local shape feature fusion for improved matching, pose estimation and 3D object recognition
title_short Local shape feature fusion for improved matching, pose estimation and 3D object recognition
title_sort local shape feature fusion for improved matching, pose estimation and 3d object recognition
topic Research
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4783326/
https://www.ncbi.nlm.nih.gov/pubmed/27066334
http://dx.doi.org/10.1186/s40064-016-1906-1
work_keys_str_mv AT buchandersg localshapefeaturefusionforimprovedmatchingposeestimationand3dobjectrecognition
AT petersenhenrikg localshapefeaturefusionforimprovedmatchingposeestimationand3dobjectrecognition
AT krugernorbert localshapefeaturefusionforimprovedmatchingposeestimationand3dobjectrecognition