3-D reconstruction of multipart self-occluding objects

Nebojsa Jojic, Jin Gu, Helen C. Shen, Thomas Huang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper we present a method for reconstruction of multipart objects from several arbitrary views using deformable superquadrics as the models of the object’s parts. Two visual cues are used: occluding contours and stereo (possibly aided by projected patterns). The object can be relatively complex and can exhibit numerous self occlusions from some or all views. Our preliminary experiments on a human body and a tailor’s mannequin show that the reconstruction is more complete than in purely stereo or structured light based methods and more precise than the reconstruction from occluding contours only.

Original languageEnglish (US)
Title of host publicationComputer Vision - ACCV 1998 - 3rd Asian Conference on Computer Vision, Proceedings
EditorsRoland Chin, Ting-Chuen Pong
PublisherSpringer-Verlag
Pages455-462
Number of pages8
ISBN (Print)3540639314, 9783540639312
StatePublished - Jan 1 1998
Event3rd Asian Conference on Computer Vision, ACCV 1998 - Hong Kong, Hong Kong
Duration: Jan 8 1998Jan 10 1998

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1352
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other3rd Asian Conference on Computer Vision, ACCV 1998
CountryHong Kong
CityHong Kong
Period1/8/981/10/98

Fingerprint

3D Reconstruction
Structured Light
Experiments
Occlusion
Arbitrary
Experiment
Object
Model

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Jojic, N., Gu, J., Shen, H. C., & Huang, T. (1998). 3-D reconstruction of multipart self-occluding objects. In R. Chin, & T-C. Pong (Eds.), Computer Vision - ACCV 1998 - 3rd Asian Conference on Computer Vision, Proceedings (pp. 455-462). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1352). Springer-Verlag.

3-D reconstruction of multipart self-occluding objects. / Jojic, Nebojsa; Gu, Jin; Shen, Helen C.; Huang, Thomas.

Computer Vision - ACCV 1998 - 3rd Asian Conference on Computer Vision, Proceedings. ed. / Roland Chin; Ting-Chuen Pong. Springer-Verlag, 1998. p. 455-462 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1352).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Jojic, N, Gu, J, Shen, HC & Huang, T 1998, 3-D reconstruction of multipart self-occluding objects. in R Chin & T-C Pong (eds), Computer Vision - ACCV 1998 - 3rd Asian Conference on Computer Vision, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1352, Springer-Verlag, pp. 455-462, 3rd Asian Conference on Computer Vision, ACCV 1998, Hong Kong, Hong Kong, 1/8/98.
Jojic N, Gu J, Shen HC, Huang T. 3-D reconstruction of multipart self-occluding objects. In Chin R, Pong T-C, editors, Computer Vision - ACCV 1998 - 3rd Asian Conference on Computer Vision, Proceedings. Springer-Verlag. 1998. p. 455-462. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Jojic, Nebojsa ; Gu, Jin ; Shen, Helen C. ; Huang, Thomas. / 3-D reconstruction of multipart self-occluding objects. Computer Vision - ACCV 1998 - 3rd Asian Conference on Computer Vision, Proceedings. editor / Roland Chin ; Ting-Chuen Pong. Springer-Verlag, 1998. pp. 455-462 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{05133c5d3b0c4632a6e8e8fa34f95ef9,
title = "3-D reconstruction of multipart self-occluding objects",
abstract = "In this paper we present a method for reconstruction of multipart objects from several arbitrary views using deformable superquadrics as the models of the object’s parts. Two visual cues are used: occluding contours and stereo (possibly aided by projected patterns). The object can be relatively complex and can exhibit numerous self occlusions from some or all views. Our preliminary experiments on a human body and a tailor’s mannequin show that the reconstruction is more complete than in purely stereo or structured light based methods and more precise than the reconstruction from occluding contours only.",
author = "Nebojsa Jojic and Jin Gu and Shen, {Helen C.} and Thomas Huang",
year = "1998",
month = "1",
day = "1",
language = "English (US)",
isbn = "3540639314",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag",
pages = "455--462",
editor = "Roland Chin and Ting-Chuen Pong",
booktitle = "Computer Vision - ACCV 1998 - 3rd Asian Conference on Computer Vision, Proceedings",

}

TY - GEN

T1 - 3-D reconstruction of multipart self-occluding objects

AU - Jojic, Nebojsa

AU - Gu, Jin

AU - Shen, Helen C.

AU - Huang, Thomas

PY - 1998/1/1

Y1 - 1998/1/1

N2 - In this paper we present a method for reconstruction of multipart objects from several arbitrary views using deformable superquadrics as the models of the object’s parts. Two visual cues are used: occluding contours and stereo (possibly aided by projected patterns). The object can be relatively complex and can exhibit numerous self occlusions from some or all views. Our preliminary experiments on a human body and a tailor’s mannequin show that the reconstruction is more complete than in purely stereo or structured light based methods and more precise than the reconstruction from occluding contours only.

AB - In this paper we present a method for reconstruction of multipart objects from several arbitrary views using deformable superquadrics as the models of the object’s parts. Two visual cues are used: occluding contours and stereo (possibly aided by projected patterns). The object can be relatively complex and can exhibit numerous self occlusions from some or all views. Our preliminary experiments on a human body and a tailor’s mannequin show that the reconstruction is more complete than in purely stereo or structured light based methods and more precise than the reconstruction from occluding contours only.

UR - http://www.scopus.com/inward/record.url?scp=84901468788&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901468788&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84901468788

SN - 3540639314

SN - 9783540639312

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 455

EP - 462

BT - Computer Vision - ACCV 1998 - 3rd Asian Conference on Computer Vision, Proceedings

A2 - Chin, Roland

A2 - Pong, Ting-Chuen

PB - Springer-Verlag

ER -