Simple Combination of Appearance and Depth for Foreground Segmentation

研究成果: 著書/レポートタイプへの貢献会議での発言

8 引用 (Scopus)

抄録

In foreground segmentation, the depth information is robust to problems of the appearance information such as illumination changes and color camouflage; however, the depth information is not always measured and suffers from depth camouflage. In order to compensate for the disadvantages of the two pieces of information, we define an energy function based on the two likelihoods of depth and appearance backgrounds and minimize the energy using graph cuts to obtain a foreground mask. The two likelihoods are obtained using background subtraction. We use the farthest depth as the depth background in the background subtraction according to the depth information. The appearance background is defined as the appearance with a large likelihood of the depth background to eliminate appearances of foreground objects. In the computation of the likelihood of the appearance background, we also use the likelihood of the depth background for reducing false positives owing to illumination changes. In our experiment, we confirm that our method is sufficiently accurate for indoor environments using the SBM-RGBD 2017 dataset.

元の言語英語
ホスト出版物のタイトルNew Trends in Image Analysis and Processing – ICIAP 2017 - ICIAP International Workshops, WBICV, SSPandBE, 3AS, RGBD, NIVAR, IWBAAS, and MADiMa 2017, Revised Selected Papers
編集者Sebastiano Battiato, Giovanni Maria Farinella, Marco Leo, Giovanni Gallo
出版者Springer Verlag
ページ266-277
ページ数12
ISBN(印刷物)9783319707419
DOI
出版物ステータス出版済み - 1 1 2017
イベント19th International Conference on Image Analysis and Processing, ICIAP 2017 - Catania, イタリア
継続期間: 6 5 20176 9 2017

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
10590 LNCS
ISSN(印刷物)0302-9743
ISSN(電子版)1611-3349

その他

その他19th International Conference on Image Analysis and Processing, ICIAP 2017
イタリア
Catania
期間6/5/176/9/17

Fingerprint

Camouflage
Segmentation
Lighting
Likelihood
Masks
Color
Background Subtraction
Illumination
Experiments
Graph Cuts
Energy Function
False Positive
Mask
Background
Eliminate
Minimise

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

これを引用

Minematsu, T., Shimada, A., Uchiyama, H., & Taniguchi, R-I. (2017). Simple Combination of Appearance and Depth for Foreground Segmentation. : S. Battiato, G. M. Farinella, M. Leo, & G. Gallo (版), New Trends in Image Analysis and Processing – ICIAP 2017 - ICIAP International Workshops, WBICV, SSPandBE, 3AS, RGBD, NIVAR, IWBAAS, and MADiMa 2017, Revised Selected Papers (pp. 266-277). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); 巻数 10590 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-70742-6_25

Simple Combination of Appearance and Depth for Foreground Segmentation. / Minematsu, Tsubasa; Shimada, Atsushi; Uchiyama, Hideaki; Taniguchi, Rin-Ichiro.

New Trends in Image Analysis and Processing – ICIAP 2017 - ICIAP International Workshops, WBICV, SSPandBE, 3AS, RGBD, NIVAR, IWBAAS, and MADiMa 2017, Revised Selected Papers. 版 / Sebastiano Battiato; Giovanni Maria Farinella; Marco Leo; Giovanni Gallo. Springer Verlag, 2017. p. 266-277 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); 巻 10590 LNCS).

研究成果: 著書/レポートタイプへの貢献会議での発言

Minematsu, T, Shimada, A, Uchiyama, H & Taniguchi, R-I 2017, Simple Combination of Appearance and Depth for Foreground Segmentation. : S Battiato, GM Farinella, M Leo & G Gallo (版), New Trends in Image Analysis and Processing – ICIAP 2017 - ICIAP International Workshops, WBICV, SSPandBE, 3AS, RGBD, NIVAR, IWBAAS, and MADiMa 2017, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 巻. 10590 LNCS, Springer Verlag, pp. 266-277, 19th International Conference on Image Analysis and Processing, ICIAP 2017, Catania, イタリア, 6/5/17. https://doi.org/10.1007/978-3-319-70742-6_25
Minematsu T, Shimada A, Uchiyama H, Taniguchi R-I. Simple Combination of Appearance and Depth for Foreground Segmentation. : Battiato S, Farinella GM, Leo M, Gallo G, 編集者, New Trends in Image Analysis and Processing – ICIAP 2017 - ICIAP International Workshops, WBICV, SSPandBE, 3AS, RGBD, NIVAR, IWBAAS, and MADiMa 2017, Revised Selected Papers. Springer Verlag. 2017. p. 266-277. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-70742-6_25
Minematsu, Tsubasa ; Shimada, Atsushi ; Uchiyama, Hideaki ; Taniguchi, Rin-Ichiro. / Simple Combination of Appearance and Depth for Foreground Segmentation. New Trends in Image Analysis and Processing – ICIAP 2017 - ICIAP International Workshops, WBICV, SSPandBE, 3AS, RGBD, NIVAR, IWBAAS, and MADiMa 2017, Revised Selected Papers. 編集者 / Sebastiano Battiato ; Giovanni Maria Farinella ; Marco Leo ; Giovanni Gallo. Springer Verlag, 2017. pp. 266-277 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{0b2d1edda336486bb336936381dd9c17,
title = "Simple Combination of Appearance and Depth for Foreground Segmentation",
abstract = "In foreground segmentation, the depth information is robust to problems of the appearance information such as illumination changes and color camouflage; however, the depth information is not always measured and suffers from depth camouflage. In order to compensate for the disadvantages of the two pieces of information, we define an energy function based on the two likelihoods of depth and appearance backgrounds and minimize the energy using graph cuts to obtain a foreground mask. The two likelihoods are obtained using background subtraction. We use the farthest depth as the depth background in the background subtraction according to the depth information. The appearance background is defined as the appearance with a large likelihood of the depth background to eliminate appearances of foreground objects. In the computation of the likelihood of the appearance background, we also use the likelihood of the depth background for reducing false positives owing to illumination changes. In our experiment, we confirm that our method is sufficiently accurate for indoor environments using the SBM-RGBD 2017 dataset.",
author = "Tsubasa Minematsu and Atsushi Shimada and Hideaki Uchiyama and Rin-Ichiro Taniguchi",
year = "2017",
month = "1",
day = "1",
doi = "10.1007/978-3-319-70742-6_25",
language = "English",
isbn = "9783319707419",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "266--277",
editor = "Sebastiano Battiato and Farinella, {Giovanni Maria} and Marco Leo and Giovanni Gallo",
booktitle = "New Trends in Image Analysis and Processing – ICIAP 2017 - ICIAP International Workshops, WBICV, SSPandBE, 3AS, RGBD, NIVAR, IWBAAS, and MADiMa 2017, Revised Selected Papers",
address = "Germany",

}

TY - GEN

T1 - Simple Combination of Appearance and Depth for Foreground Segmentation

AU - Minematsu, Tsubasa

AU - Shimada, Atsushi

AU - Uchiyama, Hideaki

AU - Taniguchi, Rin-Ichiro

PY - 2017/1/1

Y1 - 2017/1/1

N2 - In foreground segmentation, the depth information is robust to problems of the appearance information such as illumination changes and color camouflage; however, the depth information is not always measured and suffers from depth camouflage. In order to compensate for the disadvantages of the two pieces of information, we define an energy function based on the two likelihoods of depth and appearance backgrounds and minimize the energy using graph cuts to obtain a foreground mask. The two likelihoods are obtained using background subtraction. We use the farthest depth as the depth background in the background subtraction according to the depth information. The appearance background is defined as the appearance with a large likelihood of the depth background to eliminate appearances of foreground objects. In the computation of the likelihood of the appearance background, we also use the likelihood of the depth background for reducing false positives owing to illumination changes. In our experiment, we confirm that our method is sufficiently accurate for indoor environments using the SBM-RGBD 2017 dataset.

AB - In foreground segmentation, the depth information is robust to problems of the appearance information such as illumination changes and color camouflage; however, the depth information is not always measured and suffers from depth camouflage. In order to compensate for the disadvantages of the two pieces of information, we define an energy function based on the two likelihoods of depth and appearance backgrounds and minimize the energy using graph cuts to obtain a foreground mask. The two likelihoods are obtained using background subtraction. We use the farthest depth as the depth background in the background subtraction according to the depth information. The appearance background is defined as the appearance with a large likelihood of the depth background to eliminate appearances of foreground objects. In the computation of the likelihood of the appearance background, we also use the likelihood of the depth background for reducing false positives owing to illumination changes. In our experiment, we confirm that our method is sufficiently accurate for indoor environments using the SBM-RGBD 2017 dataset.

UR - http://www.scopus.com/inward/record.url?scp=85041129216&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041129216&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-70742-6_25

DO - 10.1007/978-3-319-70742-6_25

M3 - Conference contribution

AN - SCOPUS:85041129216

SN - 9783319707419

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

SP - 266

EP - 277

BT - New Trends in Image Analysis and Processing – ICIAP 2017 - ICIAP International Workshops, WBICV, SSPandBE, 3AS, RGBD, NIVAR, IWBAAS, and MADiMa 2017, Revised Selected Papers

A2 - Battiato, Sebastiano

A2 - Farinella, Giovanni Maria

A2 - Leo, Marco

A2 - Gallo, Giovanni

PB - Springer Verlag

ER -