Profile hidden Markov models for foreground object modelling

Ioannis Kazantzidis; Francisco Florez Revuelta; Jean-Christophe Nebel

doi:10.1109/ICIP.2018.8451476

Profile hidden Markov models for foreground object modelling

Ioannis Kazantzidis
, Francisco Florez Revuelta
, Jean-Christophe Nebel

Research output: Contribution to conference › Paper › peer-review

3 Downloads (Pure)

Abstract

Accurate background/foreground segmentation is a preliminary process essential to most visual surveillance applications. With the increasing use of freely moving cameras, strategies have been proposed to refine initial segmentation. In this paper, it is proposed to exploit the Vide-omics paradigm, and Profile Hidden Markov Models in particular, to create a new type of object descriptors relying on spatiotemporal information. Performance of the proposed methodology has been evaluated using a standard dataset of videos captured by moving cameras. Results show that usage of the proposed object descriptors allows better foreground extraction than standard approaches.

Original language	English
DOIs	https://doi.org/10.1109/ICIP.2018.8451476
Publication status	Published - Oct 2018
Event	25th IEEE International Conference on Image Processing (ICIP) - Athens, Greece Duration: 7 Oct 2018 → 10 Oct 2018

Conference

Conference	25th IEEE International Conference on Image Processing (ICIP)
Period	7/10/18 → 10/10/18

Bibliographical note

Note: Published in: 2018 25th IEEE International Conference on Image Processing (ICIP). Piscataway, U.S. : Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2381-8549 ISBN 9781479970629

Organising Body: IEEE

Keywords

Computer science and informatics

Access to Document

10.1109/ICIP.2018.8451476

Kazantzidis-I-42430-AAMAccepted author manuscript, 402 KBLicence: Unspecified

http://dx.doi.org/10.1109/ICIP.2018.8451476

1 Finished

Vide-omics: A Genomics-inspired Paradigm for Video Analysis
Nebel, J.-C. (PI), Kazantzidis, I. (Researcher), Florez-Revuelta, F. (CoI), Del Rincon, J. M. (CoI), Thevenon, J. (Researcher), Dieny, R. (Researcher), Dequidt, M. (Researcher), Hill, N. (CoI) & Dos-Santos-Paulino, A. C. (Researcher)
1/09/10 → 30/10/18
Project: Research

1 Conference contribution

Profile hidden Markov models for foreground object modelling
Kazantzidis, I., Florez-Revuelta, F. & Nebel, J. C., Oct 2018, Published in: 2018 25th IEEE International Conference on Image Processing (ICIP). Piscataway, U.S. : Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2381-8549 ISBN 9781479970629 Organising Body: IEEE Organising Body: IEEE.
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Open Access
File

Cite this

@conference{110f0df905654f6c943c98fe57e1f8c4,

title = "Profile hidden Markov models for foreground object modelling",

abstract = "Accurate background/foreground segmentation is a preliminary process essential to most visual surveillance applications. With the increasing use of freely moving cameras, strategies have been proposed to refine initial segmentation. In this paper, it is proposed to exploit the Vide-omics paradigm, and Profile Hidden Markov Models in particular, to create a new type of object descriptors relying on spatiotemporal information. Performance of the proposed methodology has been evaluated using a standard dataset of videos captured by moving cameras. Results show that usage of the proposed object descriptors allows better foreground extraction than standard approaches.",

keywords = "Computer science and informatics",

author = "Ioannis Kazantzidis and \{Florez Revuelta\}, Francisco and Jean-Christophe Nebel",

note = "Note: Published in: 2018 25th IEEE International Conference on Image Processing (ICIP). Piscataway, U.S. : Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2381-8549 ISBN 9781479970629 Organising Body: IEEE; 25th IEEE International Conference on Image Processing (ICIP) ; Conference date: 07-10-2018 Through 10-10-2018",

year = "2018",

month = oct,

doi = "10.1109/ICIP.2018.8451476",

language = "English",

}

TY - CONF

T1 - Profile hidden Markov models for foreground object modelling

AU - Kazantzidis, Ioannis

AU - Florez Revuelta, Francisco

AU - Nebel, Jean-Christophe

N1 - Note: Published in: 2018 25th IEEE International Conference on Image Processing (ICIP). Piscataway, U.S. : Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2381-8549 ISBN 9781479970629 Organising Body: IEEE

PY - 2018/10

Y1 - 2018/10

N2 - Accurate background/foreground segmentation is a preliminary process essential to most visual surveillance applications. With the increasing use of freely moving cameras, strategies have been proposed to refine initial segmentation. In this paper, it is proposed to exploit the Vide-omics paradigm, and Profile Hidden Markov Models in particular, to create a new type of object descriptors relying on spatiotemporal information. Performance of the proposed methodology has been evaluated using a standard dataset of videos captured by moving cameras. Results show that usage of the proposed object descriptors allows better foreground extraction than standard approaches.

AB - Accurate background/foreground segmentation is a preliminary process essential to most visual surveillance applications. With the increasing use of freely moving cameras, strategies have been proposed to refine initial segmentation. In this paper, it is proposed to exploit the Vide-omics paradigm, and Profile Hidden Markov Models in particular, to create a new type of object descriptors relying on spatiotemporal information. Performance of the proposed methodology has been evaluated using a standard dataset of videos captured by moving cameras. Results show that usage of the proposed object descriptors allows better foreground extraction than standard approaches.

KW - Computer science and informatics

U2 - 10.1109/ICIP.2018.8451476

DO - 10.1109/ICIP.2018.8451476

M3 - Paper

T2 - 25th IEEE International Conference on Image Processing (ICIP)

Y2 - 7 October 2018 through 10 October 2018

ER -

Profile hidden Markov models for foreground object modelling

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Projects

Vide-omics: A Genomics-inspired Paradigm for Video Analysis

Research output

Profile hidden Markov models for foreground object modelling

Cite this