Nonlinear PCA/ICA for the structure from motion problem

Jun Fujiki; Shotaro Akaho; Noboru Murata

doi:10.1007/978-3-540-30110-3_95

Nonlinear PCA/ICA for the structure from motion problem

Jun Fujiki^*, Shotaro Akaho, Noboru Murata

^*Corresponding author for this work

School of Advanced Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Chapter

2 Citations (Scopus)

Abstract

Recovering both camera motion and object shape from multiple images,' called structure from motion problem, is an important and essential problem in computer vision. Generally, the result of the structure from motion problem has an ambiguity represented by a threedimensional rotation matrix. We present two kinds of specific criteria such as independence of parameters to fix the ambiguity by choosing an appropriate rotation matrix in the sense of computer vision. Once some criterion is defined, the fixing of the ambiguity is reduced to a nonlinear extension of the PCA/ICA. We examine the efficiency through synthetic experiments.

Original language	English
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Editors	Carlos G. Puntonet, Alberto Prieto
Publisher	Springer Verlag
Pages	750-757
Number of pages	8
ISBN (Electronic)	3540230564, 9783540230564
DOIs	https://doi.org/10.1007/978-3-540-30110-3_95
Publication status	Published - 2004 Jan 1

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	3195
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-540-30110-3_95

Cite this

Fujiki, J., Akaho, S., & Murata, N. (2004). Nonlinear PCA/ICA for the structure from motion problem. In C. G. Puntonet, & A. Prieto (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 750-757). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3195). Springer Verlag. https://doi.org/10.1007/978-3-540-30110-3_95

Nonlinear PCA/ICA for the structure from motion problem. / Fujiki, Jun; Akaho, Shotaro; Murata, Noboru.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). ed. / Carlos G. Puntonet; Alberto Prieto. Springer Verlag, 2004. p. 750-757 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3195).

Research output: Chapter in Book/Report/Conference proceeding › Chapter

Fujiki, J, Akaho, S & Murata, N 2004, Nonlinear PCA/ICA for the structure from motion problem. in CG Puntonet & A Prieto (eds), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3195, Springer Verlag, pp. 750-757. https://doi.org/10.1007/978-3-540-30110-3_95

Fujiki J, Akaho S, Murata N. Nonlinear PCA/ICA for the structure from motion problem. In Puntonet CG, Prieto A, editors, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer Verlag. 2004. p. 750-757. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-540-30110-3_95

Fujiki, Jun ; Akaho, Shotaro ; Murata, Noboru. / Nonlinear PCA/ICA for the structure from motion problem. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). editor / Carlos G. Puntonet ; Alberto Prieto. Springer Verlag, 2004. pp. 750-757 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inbook{5f45b4b72a6d4add983c64052aca8a56,

title = "Nonlinear PCA/ICA for the structure from motion problem",

abstract = "Recovering both camera motion and object shape from multiple images,' called structure from motion problem, is an important and essential problem in computer vision. Generally, the result of the structure from motion problem has an ambiguity represented by a threedimensional rotation matrix. We present two kinds of specific criteria such as independence of parameters to fix the ambiguity by choosing an appropriate rotation matrix in the sense of computer vision. Once some criterion is defined, the fixing of the ambiguity is reduced to a nonlinear extension of the PCA/ICA. We examine the efficiency through synthetic experiments.",

author = "Jun Fujiki and Shotaro Akaho and Noboru Murata",

year = "2004",

month = jan,

day = "1",

doi = "10.1007/978-3-540-30110-3_95",

language = "English",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "750--757",

editor = "Puntonet, {Carlos G.} and Alberto Prieto",

booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - CHAP

T1 - Nonlinear PCA/ICA for the structure from motion problem

AU - Fujiki, Jun

AU - Akaho, Shotaro

AU - Murata, Noboru

PY - 2004/1/1

Y1 - 2004/1/1

N2 - Recovering both camera motion and object shape from multiple images,' called structure from motion problem, is an important and essential problem in computer vision. Generally, the result of the structure from motion problem has an ambiguity represented by a threedimensional rotation matrix. We present two kinds of specific criteria such as independence of parameters to fix the ambiguity by choosing an appropriate rotation matrix in the sense of computer vision. Once some criterion is defined, the fixing of the ambiguity is reduced to a nonlinear extension of the PCA/ICA. We examine the efficiency through synthetic experiments.

AB - Recovering both camera motion and object shape from multiple images,' called structure from motion problem, is an important and essential problem in computer vision. Generally, the result of the structure from motion problem has an ambiguity represented by a threedimensional rotation matrix. We present two kinds of specific criteria such as independence of parameters to fix the ambiguity by choosing an appropriate rotation matrix in the sense of computer vision. Once some criterion is defined, the fixing of the ambiguity is reduced to a nonlinear extension of the PCA/ICA. We examine the efficiency through synthetic experiments.

UR - http://www.scopus.com/inward/record.url?scp=35048843747&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35048843747&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-30110-3_95

DO - 10.1007/978-3-540-30110-3_95

M3 - Chapter

AN - SCOPUS:35048843747

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

SP - 750

EP - 757

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

A2 - Puntonet, Carlos G.

A2 - Prieto, Alberto

PB - Springer Verlag

ER -

Nonlinear PCA/ICA for the structure from motion problem

Abstract

Publication series

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this