Robust approach for interesting points extraction of moving human from 2D videos

Yu Xiang, Yoshihiro Okada

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Human action recognition is a key technique for content-based video retrieval. Because a human motion consists of several sequential poses of the human, specifying each poses of the human motion is required for human action recognition. In this paper, the authors focus on end points and joint points of a human skeleton as interesting points obtainable from the human silhouette image of each video frame including a human motion because those points are important for specifying human poses in the human motion. This paper presents a stable and effective end points and joint points extraction method for the human body from 2D videos. The authors employ a perfect foreground object segmentation algorithm by background subtraction to obtain a moving object. Morphological and connection labeling-based algorithms are then performed on foreground objects. In addition, the paper considers the cast shadow and skeleton pruning problem which will influence the accuracy of the interesting points extraction. The experiments also show the good results of the proposed method.

Original languageEnglish
Title of host publicationInformation Search, Integration, and Personalization - 10th International Workshop, ISIP 2015, Revised Selected Papers
EditorsYuzuru Tanaka, Emanuel Grant, Dimitris Kotzinos, Dominique Laurent, Nicolas Spyratos
PublisherSpringer Verlag
Pages88-104
Number of pages17
ISBN (Print)9783319438610
DOIs
Publication statusPublished - Jan 1 2016
Event10th International Workshop on Information Search, Integration, and Personalization, ISIP 2015 - Grand Forks, United States
Duration: Oct 1 2015Oct 2 2015

Publication series

NameCommunications in Computer and Information Science
Volume622
ISSN (Print)1865-0929

Other

Other10th International Workshop on Information Search, Integration, and Personalization, ISIP 2015
CountryUnited States
CityGrand Forks
Period10/1/1510/2/15

Fingerprint

Labeling
Action Recognition
Motion
End point
Experiments
Skeleton
Human
Video Retrieval
Content-based Retrieval
Background Subtraction
Silhouette
Pruning
Moving Objects
Segmentation
Experiment
Object

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Mathematics(all)

Cite this

Xiang, Y., & Okada, Y. (2016). Robust approach for interesting points extraction of moving human from 2D videos. In Y. Tanaka, E. Grant, D. Kotzinos, D. Laurent, & N. Spyratos (Eds.), Information Search, Integration, and Personalization - 10th International Workshop, ISIP 2015, Revised Selected Papers (pp. 88-104). (Communications in Computer and Information Science; Vol. 622). Springer Verlag. https://doi.org/10.1007/978-3-319-43862-7_5

Robust approach for interesting points extraction of moving human from 2D videos. / Xiang, Yu; Okada, Yoshihiro.

Information Search, Integration, and Personalization - 10th International Workshop, ISIP 2015, Revised Selected Papers. ed. / Yuzuru Tanaka; Emanuel Grant; Dimitris Kotzinos; Dominique Laurent; Nicolas Spyratos. Springer Verlag, 2016. p. 88-104 (Communications in Computer and Information Science; Vol. 622).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Xiang, Y & Okada, Y 2016, Robust approach for interesting points extraction of moving human from 2D videos. in Y Tanaka, E Grant, D Kotzinos, D Laurent & N Spyratos (eds), Information Search, Integration, and Personalization - 10th International Workshop, ISIP 2015, Revised Selected Papers. Communications in Computer and Information Science, vol. 622, Springer Verlag, pp. 88-104, 10th International Workshop on Information Search, Integration, and Personalization, ISIP 2015, Grand Forks, United States, 10/1/15. https://doi.org/10.1007/978-3-319-43862-7_5
Xiang Y, Okada Y. Robust approach for interesting points extraction of moving human from 2D videos. In Tanaka Y, Grant E, Kotzinos D, Laurent D, Spyratos N, editors, Information Search, Integration, and Personalization - 10th International Workshop, ISIP 2015, Revised Selected Papers. Springer Verlag. 2016. p. 88-104. (Communications in Computer and Information Science). https://doi.org/10.1007/978-3-319-43862-7_5
Xiang, Yu ; Okada, Yoshihiro. / Robust approach for interesting points extraction of moving human from 2D videos. Information Search, Integration, and Personalization - 10th International Workshop, ISIP 2015, Revised Selected Papers. editor / Yuzuru Tanaka ; Emanuel Grant ; Dimitris Kotzinos ; Dominique Laurent ; Nicolas Spyratos. Springer Verlag, 2016. pp. 88-104 (Communications in Computer and Information Science).
@inproceedings{a15c12e6067b4fe88ac173ef7a4ef462,
title = "Robust approach for interesting points extraction of moving human from 2D videos",
abstract = "Human action recognition is a key technique for content-based video retrieval. Because a human motion consists of several sequential poses of the human, specifying each poses of the human motion is required for human action recognition. In this paper, the authors focus on end points and joint points of a human skeleton as interesting points obtainable from the human silhouette image of each video frame including a human motion because those points are important for specifying human poses in the human motion. This paper presents a stable and effective end points and joint points extraction method for the human body from 2D videos. The authors employ a perfect foreground object segmentation algorithm by background subtraction to obtain a moving object. Morphological and connection labeling-based algorithms are then performed on foreground objects. In addition, the paper considers the cast shadow and skeleton pruning problem which will influence the accuracy of the interesting points extraction. The experiments also show the good results of the proposed method.",
author = "Yu Xiang and Yoshihiro Okada",
year = "2016",
month = "1",
day = "1",
doi = "10.1007/978-3-319-43862-7_5",
language = "English",
isbn = "9783319438610",
series = "Communications in Computer and Information Science",
publisher = "Springer Verlag",
pages = "88--104",
editor = "Yuzuru Tanaka and Emanuel Grant and Dimitris Kotzinos and Dominique Laurent and Nicolas Spyratos",
booktitle = "Information Search, Integration, and Personalization - 10th International Workshop, ISIP 2015, Revised Selected Papers",
address = "Germany",

}

TY - GEN

T1 - Robust approach for interesting points extraction of moving human from 2D videos

AU - Xiang, Yu

AU - Okada, Yoshihiro

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Human action recognition is a key technique for content-based video retrieval. Because a human motion consists of several sequential poses of the human, specifying each poses of the human motion is required for human action recognition. In this paper, the authors focus on end points and joint points of a human skeleton as interesting points obtainable from the human silhouette image of each video frame including a human motion because those points are important for specifying human poses in the human motion. This paper presents a stable and effective end points and joint points extraction method for the human body from 2D videos. The authors employ a perfect foreground object segmentation algorithm by background subtraction to obtain a moving object. Morphological and connection labeling-based algorithms are then performed on foreground objects. In addition, the paper considers the cast shadow and skeleton pruning problem which will influence the accuracy of the interesting points extraction. The experiments also show the good results of the proposed method.

AB - Human action recognition is a key technique for content-based video retrieval. Because a human motion consists of several sequential poses of the human, specifying each poses of the human motion is required for human action recognition. In this paper, the authors focus on end points and joint points of a human skeleton as interesting points obtainable from the human silhouette image of each video frame including a human motion because those points are important for specifying human poses in the human motion. This paper presents a stable and effective end points and joint points extraction method for the human body from 2D videos. The authors employ a perfect foreground object segmentation algorithm by background subtraction to obtain a moving object. Morphological and connection labeling-based algorithms are then performed on foreground objects. In addition, the paper considers the cast shadow and skeleton pruning problem which will influence the accuracy of the interesting points extraction. The experiments also show the good results of the proposed method.

UR - http://www.scopus.com/inward/record.url?scp=84981266386&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84981266386&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-43862-7_5

DO - 10.1007/978-3-319-43862-7_5

M3 - Conference contribution

SN - 9783319438610

T3 - Communications in Computer and Information Science

SP - 88

EP - 104

BT - Information Search, Integration, and Personalization - 10th International Workshop, ISIP 2015, Revised Selected Papers

A2 - Tanaka, Yuzuru

A2 - Grant, Emanuel

A2 - Kotzinos, Dimitris

A2 - Laurent, Dominique

A2 - Spyratos, Nicolas

PB - Springer Verlag

ER -