Perceptually Relevant and Piecewise Linear Matching of Silhouettes

Xenophon Zabulis; Jon Sporring; Xenophon Orphanoudakis

doi:doi:10.1016/j.patcog.2004.06.003

Perceptually Relevant and Piecewise Linear Matching of Silhouettes

Xenophon Zabulis, Jon Sporring, Xenophon Orphanoudakis

14 Citations (Scopus)

1084 Downloads (Pure)

Abstract

In this paper, a novel alignment method for silhouettes is proposed. This method is based on the establishment of correspondences between landmarks on their boundaries and, in turn, on the establishment of correspondences of the boundary pieces in between these landmarks. The method yields more correct correspondences than conventional methods that scale the arc-length descriptions of silhouettes to align them. The selection of landmarks is investigated as to the robustness of their localization and their perceptual relevance. Matching of silhouettes is then achieved by quantifying the dissimilarity of a pair of silhouette boundaries, based on a novel dissimilarity metric. The matching procedure is evaluated, based on retrieval experiments, and it is concluded that the precision of the results is higher than that obtained by conventional pointwise comparison methods.

Original language	English
Journal	Pattern Recognition
Volume	38
Issue number	1
Pages (from-to)	75-93
ISSN	0031-3203
DOIs	https://doi.org/doi:10.1016/j.patcog.2004.06.003
Publication status	Published - 2005

Access to Document

doi:10.1016/j.patcog.2004.06.003

B4Submitted manuscript, 1.75 MB

http://www.sciencedirect.com/science?_ob=MImg&_imagekey=B6V14-4D5JSK2-2-Y&_cdi=5664&_user=641710&_orig=search&_coverDate=01%2F01%2F2005&_sk=999619998&view=c&wchp=dGLbVlz-zSkWb&md5=03339e549af64af6de2c97d6d1a737e2&ie=/sdarticle.pdf

Cite this

@article{2f962c9074c211dbbee902004c4f4f50,

title = "Perceptually Relevant and Piecewise Linear Matching of Silhouettes",

abstract = "In this paper, a novel alignment method for silhouettes is proposed. This method is based on the establishment of correspondences between landmarks on their boundaries and, in turn, on the establishment of correspondences of the boundary pieces in between these landmarks. The method yields more correct correspondences than conventional methods that scale the arc-length descriptions of silhouettes to align them. The selection of landmarks is investigated as to the robustness of their localization and their perceptual relevance. Matching of silhouettes is then achieved by quantifying the dissimilarity of a pair of silhouette boundaries, based on a novel dissimilarity metric. The matching procedure is evaluated, based on retrieval experiments, and it is concluded that the precision of the results is higher than that obtained by conventional pointwise comparison methods.",

author = "Xenophon Zabulis and Jon Sporring and Xenophon Orphanoudakis",

year = "2005",

doi = "doi:10.1016/j.patcog.2004.06.003",

language = "English",

volume = "38",

pages = "75--93",

journal = "Pattern Recognition",

issn = "0031-3203",

publisher = "Elsevier",

number = "1",

}

TY - JOUR

T1 - Perceptually Relevant and Piecewise Linear Matching of Silhouettes

AU - Zabulis, Xenophon

AU - Sporring, Jon

AU - Orphanoudakis, Xenophon

PY - 2005

Y1 - 2005

N2 - In this paper, a novel alignment method for silhouettes is proposed. This method is based on the establishment of correspondences between landmarks on their boundaries and, in turn, on the establishment of correspondences of the boundary pieces in between these landmarks. The method yields more correct correspondences than conventional methods that scale the arc-length descriptions of silhouettes to align them. The selection of landmarks is investigated as to the robustness of their localization and their perceptual relevance. Matching of silhouettes is then achieved by quantifying the dissimilarity of a pair of silhouette boundaries, based on a novel dissimilarity metric. The matching procedure is evaluated, based on retrieval experiments, and it is concluded that the precision of the results is higher than that obtained by conventional pointwise comparison methods.

AB - In this paper, a novel alignment method for silhouettes is proposed. This method is based on the establishment of correspondences between landmarks on their boundaries and, in turn, on the establishment of correspondences of the boundary pieces in between these landmarks. The method yields more correct correspondences than conventional methods that scale the arc-length descriptions of silhouettes to align them. The selection of landmarks is investigated as to the robustness of their localization and their perceptual relevance. Matching of silhouettes is then achieved by quantifying the dissimilarity of a pair of silhouette boundaries, based on a novel dissimilarity metric. The matching procedure is evaluated, based on retrieval experiments, and it is concluded that the precision of the results is higher than that obtained by conventional pointwise comparison methods.

U2 - doi:10.1016/j.patcog.2004.06.003

DO - doi:10.1016/j.patcog.2004.06.003

M3 - Journal article

SN - 0031-3203

VL - 38

SP - 75

EP - 93

JO - Pattern Recognition

JF - Pattern Recognition

IS - 1

ER -

Perceptually Relevant and Piecewise Linear Matching of Silhouettes

Abstract

Access to Document

Fingerprint

Cite this