Entropy-based correspondence improvement of interpolated skeletal models

Liyun Tu, Jared Vicory, Shireen Elhabian, Beatriz Paniagua, Juan Carlos Prieto, James N. Damon, Ross Whitaker, Martin Styner, Stephen M. Pizer

Research output: Research - peer-reviewArticle

Abstract

Statistical analysis of shape representations relies on having good correspondence across a population. Improving correspondence yields improved statistics. Point distribution models (PDMs) are often used to represent object boundaries. Skeletal representations (s-reps) model object widths and boundary directions as well as boundary positions, so they should yield better correspondence. We present two methods: one for continuously interpolating a discretely-sampled skeletal model and one for improving correspondence by using this interpolation to shift skeletal samples to new positions. The interpolation operates by an extension of the mathematics of medial structures. As with Cates’ boundary-based method, we evaluate correspondence in terms of regularity and shape-feature population entropies. Evaluation on both synthetic and real data shows that our method both improves correspondence of s-rep models fit to segmented lateral ventricles and that the combined boundary-and-skeletal PDMs implied by these optimized s-reps have better correspondence than optimized boundary PDMs.

LanguageEnglish (US)
Pages72-79
Number of pages8
JournalComputer Vision and Image Understanding
Volume151
DOIs
StatePublished - Oct 1 2016

Fingerprint

Entropy
Interpolation
Statistical methods
Statistics

Keywords

  • Correspondence
  • Skeletal models
  • Statistical shape analysis

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition

Cite this

Entropy-based correspondence improvement of interpolated skeletal models. / Tu, Liyun; Vicory, Jared; Elhabian, Shireen; Paniagua, Beatriz; Prieto, Juan Carlos; Damon, James N.; Whitaker, Ross; Styner, Martin; Pizer, Stephen M.

In: Computer Vision and Image Understanding, Vol. 151, 01.10.2016, p. 72-79.

Research output: Research - peer-reviewArticle

@article{c8a816015b9646448ead417748000026,
title = "Entropy-based correspondence improvement of interpolated skeletal models",
abstract = "Statistical analysis of shape representations relies on having good correspondence across a population. Improving correspondence yields improved statistics. Point distribution models (PDMs) are often used to represent object boundaries. Skeletal representations (s-reps) model object widths and boundary directions as well as boundary positions, so they should yield better correspondence. We present two methods: one for continuously interpolating a discretely-sampled skeletal model and one for improving correspondence by using this interpolation to shift skeletal samples to new positions. The interpolation operates by an extension of the mathematics of medial structures. As with Cates’ boundary-based method, we evaluate correspondence in terms of regularity and shape-feature population entropies. Evaluation on both synthetic and real data shows that our method both improves correspondence of s-rep models fit to segmented lateral ventricles and that the combined boundary-and-skeletal PDMs implied by these optimized s-reps have better correspondence than optimized boundary PDMs.",
keywords = "Correspondence, Skeletal models, Statistical shape analysis",
author = "Liyun Tu and Jared Vicory and Shireen Elhabian and Beatriz Paniagua and Prieto, {Juan Carlos} and Damon, {James N.} and Ross Whitaker and Martin Styner and Pizer, {Stephen M.}",
year = "2016",
month = "10",
doi = "10.1016/j.cviu.2015.11.002",
volume = "151",
pages = "72--79",
journal = "Computer Vision and Image Understanding",
issn = "1077-3142",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Entropy-based correspondence improvement of interpolated skeletal models

AU - Tu,Liyun

AU - Vicory,Jared

AU - Elhabian,Shireen

AU - Paniagua,Beatriz

AU - Prieto,Juan Carlos

AU - Damon,James N.

AU - Whitaker,Ross

AU - Styner,Martin

AU - Pizer,Stephen M.

PY - 2016/10/1

Y1 - 2016/10/1

N2 - Statistical analysis of shape representations relies on having good correspondence across a population. Improving correspondence yields improved statistics. Point distribution models (PDMs) are often used to represent object boundaries. Skeletal representations (s-reps) model object widths and boundary directions as well as boundary positions, so they should yield better correspondence. We present two methods: one for continuously interpolating a discretely-sampled skeletal model and one for improving correspondence by using this interpolation to shift skeletal samples to new positions. The interpolation operates by an extension of the mathematics of medial structures. As with Cates’ boundary-based method, we evaluate correspondence in terms of regularity and shape-feature population entropies. Evaluation on both synthetic and real data shows that our method both improves correspondence of s-rep models fit to segmented lateral ventricles and that the combined boundary-and-skeletal PDMs implied by these optimized s-reps have better correspondence than optimized boundary PDMs.

AB - Statistical analysis of shape representations relies on having good correspondence across a population. Improving correspondence yields improved statistics. Point distribution models (PDMs) are often used to represent object boundaries. Skeletal representations (s-reps) model object widths and boundary directions as well as boundary positions, so they should yield better correspondence. We present two methods: one for continuously interpolating a discretely-sampled skeletal model and one for improving correspondence by using this interpolation to shift skeletal samples to new positions. The interpolation operates by an extension of the mathematics of medial structures. As with Cates’ boundary-based method, we evaluate correspondence in terms of regularity and shape-feature population entropies. Evaluation on both synthetic and real data shows that our method both improves correspondence of s-rep models fit to segmented lateral ventricles and that the combined boundary-and-skeletal PDMs implied by these optimized s-reps have better correspondence than optimized boundary PDMs.

KW - Correspondence

KW - Skeletal models

KW - Statistical shape analysis

UR - http://www.scopus.com/inward/record.url?scp=84991512295&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84991512295&partnerID=8YFLogxK

U2 - 10.1016/j.cviu.2015.11.002

DO - 10.1016/j.cviu.2015.11.002

M3 - Article

VL - 151

SP - 72

EP - 79

JO - Computer Vision and Image Understanding

T2 - Computer Vision and Image Understanding

JF - Computer Vision and Image Understanding

SN - 1077-3142

ER -