A variational approach for multi-valued velocity field estimation in transparent sequences

Alonso Ramírez-Manzanares; Mariano Rivera; Pierre Kornprobst; François Lauze

doi:10.1007/978-3-540-72823-8_20

A variational approach for multi-valued velocity field estimation in transparent sequences

Alonso Ramírez-Manzanares^*, Mariano Rivera, Pierre Kornprobst, François Lauze

^*Corresponding author for this work

Image Analysis, Computational Modelling and Geometry

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

4 Citations (Scopus)

Abstract

We propose a variational approach for multi-valued velocity field estimation in transparent sequences. Starting from existing local motion estimators, we show a variational model for integrating in space and time these local estimations to obtain a robust estimation of the multi-valued velocity field. With this approach, we can indeed estimate some multi-valued velocity fields which are not necessarily piecewise constant on a layer: Each layer can evolve according to non-parametric optical flow. We show how our approach outperforms some existing approaches, and we illustrate its capabilities on several challenging synthetic/real sequences.

Original language	English
Title of host publication	Scale Space and Variational Methods in Computer Vision, First International Conference, SSVM 2007, Proceedings
Number of pages	12
Publisher	Springer Verlag
Publication date	2007
Pages	227-238
ISBN (Print)	9783540728221
DOIs	https://doi.org/10.1007/978-3-540-72823-8_20
Publication status	Published - 2007
Event	1st International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2007 - Ischia, Italy Duration: 30 May 2007 → 2 Jun 2007

Conference

Conference	1st International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2007
Country/Territory	Italy
City	Ischia
Period	30/05/2007 → 02/06/2007

Series	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	4485 LNCS
ISSN	0302-9743

Keywords

Model competition
Multi-valued velocity fields
Transparent motion
Variational approaches

Access to Document

10.1007/978-3-540-72823-8_20

Cite this

Ramírez-Manzanares, A., Rivera, M., Kornprobst, P., & Lauze, F. (2007). A variational approach for multi-valued velocity field estimation in transparent sequences. In Scale Space and Variational Methods in Computer Vision, First International Conference, SSVM 2007, Proceedings (pp. 227-238). Springer Verlag. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 4485 LNCS https://doi.org/10.1007/978-3-540-72823-8_20

A variational approach for multi-valued velocity field estimation in transparent sequences. / Ramírez-Manzanares, Alonso; Rivera, Mariano; Kornprobst, Pierre; Lauze, François.

Scale Space and Variational Methods in Computer Vision, First International Conference, SSVM 2007, Proceedings. Springer Verlag, 2007. p. 227-238 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 4485 LNCS).

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

Ramírez-Manzanares, A, Rivera, M, Kornprobst, P & Lauze, F 2007, A variational approach for multi-valued velocity field estimation in transparent sequences. in Scale Space and Variational Methods in Computer Vision, First International Conference, SSVM 2007, Proceedings. Springer Verlag, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 4485 LNCS, pp. 227-238, 1st International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2007, Ischia, Italy, 30/05/2007. https://doi.org/10.1007/978-3-540-72823-8_20

Ramírez-Manzanares A, Rivera M, Kornprobst P, Lauze F. A variational approach for multi-valued velocity field estimation in transparent sequences. In Scale Space and Variational Methods in Computer Vision, First International Conference, SSVM 2007, Proceedings. Springer Verlag. 2007. p. 227-238. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 4485 LNCS). https://doi.org/10.1007/978-3-540-72823-8_20

Ramírez-Manzanares, Alonso ; Rivera, Mariano ; Kornprobst, Pierre ; Lauze, François. / A variational approach for multi-valued velocity field estimation in transparent sequences. Scale Space and Variational Methods in Computer Vision, First International Conference, SSVM 2007, Proceedings. Springer Verlag, 2007. pp. 227-238 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 4485 LNCS).

@inproceedings{cf627c0a64ff487f9078e32a790f7c2f,

title = "A variational approach for multi-valued velocity field estimation in transparent sequences",

abstract = "We propose a variational approach for multi-valued velocity field estimation in transparent sequences. Starting from existing local motion estimators, we show a variational model for integrating in space and time these local estimations to obtain a robust estimation of the multi-valued velocity field. With this approach, we can indeed estimate some multi-valued velocity fields which are not necessarily piecewise constant on a layer: Each layer can evolve according to non-parametric optical flow. We show how our approach outperforms some existing approaches, and we illustrate its capabilities on several challenging synthetic/real sequences.",

keywords = "Model competition, Multi-valued velocity fields, Transparent motion, Variational approaches",

author = "Alonso Ram{\'i}rez-Manzanares and Mariano Rivera and Pierre Kornprobst and Fran{\c c}ois Lauze",

year = "2007",

doi = "10.1007/978-3-540-72823-8_20",

language = "English",

isbn = "9783540728221",

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

publisher = "Springer Verlag",

pages = "227--238",

booktitle = "Scale Space and Variational Methods in Computer Vision, First International Conference, SSVM 2007, Proceedings",

note = "1st International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2007 ; Conference date: 30-05-2007 Through 02-06-2007",

}

TY - GEN

T1 - A variational approach for multi-valued velocity field estimation in transparent sequences

AU - Ramírez-Manzanares, Alonso

AU - Rivera, Mariano

AU - Kornprobst, Pierre

AU - Lauze, François

PY - 2007

Y1 - 2007

N2 - We propose a variational approach for multi-valued velocity field estimation in transparent sequences. Starting from existing local motion estimators, we show a variational model for integrating in space and time these local estimations to obtain a robust estimation of the multi-valued velocity field. With this approach, we can indeed estimate some multi-valued velocity fields which are not necessarily piecewise constant on a layer: Each layer can evolve according to non-parametric optical flow. We show how our approach outperforms some existing approaches, and we illustrate its capabilities on several challenging synthetic/real sequences.

AB - We propose a variational approach for multi-valued velocity field estimation in transparent sequences. Starting from existing local motion estimators, we show a variational model for integrating in space and time these local estimations to obtain a robust estimation of the multi-valued velocity field. With this approach, we can indeed estimate some multi-valued velocity fields which are not necessarily piecewise constant on a layer: Each layer can evolve according to non-parametric optical flow. We show how our approach outperforms some existing approaches, and we illustrate its capabilities on several challenging synthetic/real sequences.

KW - Model competition

KW - Multi-valued velocity fields

KW - Transparent motion

KW - Variational approaches

UR - http://www.scopus.com/inward/record.url?scp=37249044860&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-72823-8_20

DO - 10.1007/978-3-540-72823-8_20

M3 - Article in proceedings

AN - SCOPUS:37249044860

SN - 9783540728221

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

SP - 227

EP - 238

BT - Scale Space and Variational Methods in Computer Vision, First International Conference, SSVM 2007, Proceedings

PB - Springer Verlag

T2 - 1st International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2007

Y2 - 30 May 2007 through 2 June 2007

ER -

A variational approach for multi-valued velocity field estimation in transparent sequences

Abstract

Conference

Keywords

Access to Document

Other files and links

Cite this