Transparent-motion analysis

James R. Bergen; Peter J. Burt; Rajesh Hingorani; Shmuel Peleg

doi:10.1007/BFb0014908

Transparent-motion analysis

James R. Bergen
, Peter J. Burt
, Rajesh Hingorani
, Shmuel Peleg

The Rachel and Selim Benin School of Engineering and Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

7 Scopus citations

Abstract

A fundamental assumption made in formulating optical flow algorithms is that motion at any point in an image can be represented as a single pattern component undergoing a simple translation: even complex motion will ‘look like’ uniform displacement when viewed through a sufficiently small window. This assumption fails for a number of situations that commonly occur in real world images. For example, transparent surfaces moving past one another yield multiple motion components at a point. We propose an alternative formulation of the local motion assumption in which there may be two distinct patterns undergoing different motions within a given local analysis region. We then present an algorithm for the analysis of transparent motion.

Original language	English
Title of host publication	Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings
Editors	Olivier Faugeras
Publisher	Springer Verlag
Pages	566-569
Number of pages	4
ISBN (Print)	9783540525226
DOIs	https://doi.org/10.1007/BFb0014908
State	Published - 1990
Event	1st European Conference on Computer Vision, ECCV 1990 - Antibes, France Duration: 23 Apr 1990 → 27 Apr 1990

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	427 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	1st European Conference on Computer Vision, ECCV 1990
Country/Territory	France
City	Antibes
Period	23/04/90 → 27/04/90

Bibliographical note

Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 1990.

Access to Document

10.1007/BFb0014908

Cite this

Bergen, J. R., Burt, P. J., Hingorani, R., & Peleg, S. (1990). Transparent-motion analysis. In O. Faugeras (Ed.), Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings (pp. 566-569). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 427 LNCS). Springer Verlag. https://doi.org/10.1007/BFb0014908

Bergen, James R. ; Burt, Peter J. ; Hingorani, Rajesh et al. / Transparent-motion analysis. Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings. editor / Olivier Faugeras. Springer Verlag, 1990. pp. 566-569 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ab0ed30e150c493dbf9ff458e61ffb93,

title = "Transparent-motion analysis",

abstract = "A fundamental assumption made in formulating optical flow algorithms is that motion at any point in an image can be represented as a single pattern component undergoing a simple translation: even complex motion will {\textquoteleft}look like{\textquoteright} uniform displacement when viewed through a sufficiently small window. This assumption fails for a number of situations that commonly occur in real world images. For example, transparent surfaces moving past one another yield multiple motion components at a point. We propose an alternative formulation of the local motion assumption in which there may be two distinct patterns undergoing different motions within a given local analysis region. We then present an algorithm for the analysis of transparent motion.",

author = "Bergen, \{James R.\} and Burt, \{Peter J.\} and Rajesh Hingorani and Shmuel Peleg",

note = "Publisher Copyright: {\textcopyright} Springer-Verlag Berlin Heidelberg 1990.; 1st European Conference on Computer Vision, ECCV 1990 ; Conference date: 23-04-1990 Through 27-04-1990",

year = "1990",

doi = "10.1007/BFb0014908",

language = "אנגלית",

isbn = "9783540525226",

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

publisher = "Springer Verlag",

pages = "566--569",

editor = "Olivier Faugeras",

booktitle = "Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings",

address = "גרמניה",

}

Bergen, JR, Burt, PJ, Hingorani, R & Peleg, S 1990, Transparent-motion analysis. in O Faugeras (ed.), Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 427 LNCS, Springer Verlag, pp. 566-569, 1st European Conference on Computer Vision, ECCV 1990, Antibes, France, 23/04/90. https://doi.org/10.1007/BFb0014908

Transparent-motion analysis. / Bergen, James R.; Burt, Peter J.; Hingorani, Rajesh et al.
Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings. ed. / Olivier Faugeras. Springer Verlag, 1990. p. 566-569 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 427 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Transparent-motion analysis

AU - Bergen, James R.

AU - Burt, Peter J.

AU - Hingorani, Rajesh

AU - Peleg, Shmuel

PY - 1990

Y1 - 1990

N2 - A fundamental assumption made in formulating optical flow algorithms is that motion at any point in an image can be represented as a single pattern component undergoing a simple translation: even complex motion will ‘look like’ uniform displacement when viewed through a sufficiently small window. This assumption fails for a number of situations that commonly occur in real world images. For example, transparent surfaces moving past one another yield multiple motion components at a point. We propose an alternative formulation of the local motion assumption in which there may be two distinct patterns undergoing different motions within a given local analysis region. We then present an algorithm for the analysis of transparent motion.

AB - A fundamental assumption made in formulating optical flow algorithms is that motion at any point in an image can be represented as a single pattern component undergoing a simple translation: even complex motion will ‘look like’ uniform displacement when viewed through a sufficiently small window. This assumption fails for a number of situations that commonly occur in real world images. For example, transparent surfaces moving past one another yield multiple motion components at a point. We propose an alternative formulation of the local motion assumption in which there may be two distinct patterns undergoing different motions within a given local analysis region. We then present an algorithm for the analysis of transparent motion.

UR - https://www.scopus.com/pages/publications/0141694759

U2 - 10.1007/BFb0014908

DO - 10.1007/BFb0014908

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:0141694759

SN - 9783540525226

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

SP - 566

EP - 569

BT - Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings

A2 - Faugeras, Olivier

PB - Springer Verlag

T2 - 1st European Conference on Computer Vision, ECCV 1990

Y2 - 23 April 1990 through 27 April 1990

ER -

Bergen JR, Burt PJ, Hingorani R, Peleg S. Transparent-motion analysis. In Faugeras O, editor, Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings. Springer Verlag. 1990. p. 566-569. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/BFb0014908

Transparent-motion analysis

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this