Mining discriminative 3D Poselet for cross-view action recognition

Jiang Wang; Xiaohan Nie; Yin Xia; Ying Wu

doi:10.1109/WACV.2014.6836043

Mining discriminative 3D Poselet for cross-view action recognition

Jiang Wang, Xiaohan Nie, Yin Xia, Ying Wu

Electrical and Computer Engineering

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

2 Scopus citations

Abstract

This paper presents a novel approach to cross-view action recognition. Traditional cross-view action recognition methods typically rely on local appearance/motion features. In this paper, we take advantage of the recent developments of depth cameras to build a more discriminative cross-view action representation. In this representation, an action is characterized by the spatio-temporal configuration of 3D Poselets, which are discriminatively discovered with a novel Poselet mining algorithm and can be detected with view-invariant 3D Poselet detectors. The Kinect skeleton is employed to facilitate the 3D Poselet mining and 3D Poselet detectors learning, but the recognition is solely based on 2D video input. Extensive experiments have demonstrated that this new action representation significantly improves the accuracy and robustness for cross-view action recognition.

Original language	English (US)
Title of host publication	2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014
Publisher	IEEE Computer Society
Pages	634-639
Number of pages	6
ISBN (Print)	9781479949854
DOIs	https://doi.org/10.1109/WACV.2014.6836043
State	Published - Jan 1 2014
Event	2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014 - Steamboat Springs, CO, United States Duration: Mar 24 2014 → Mar 26 2014

Publication series

Name	2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014

Other

Other	2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014
Country	United States
City	Steamboat Springs, CO
Period	3/24/14 → 3/26/14

ASJC Scopus subject areas

Computer Science Applications
Computer Vision and Pattern Recognition

Access to Document

10.1109/WACV.2014.6836043

Fingerprint Dive into the research topics of 'Mining discriminative 3D Poselet for cross-view action recognition'. Together they form a unique fingerprint.

Cite this

Wang, J., Nie, X., Xia, Y., & Wu, Y. (2014). Mining discriminative 3D Poselet for cross-view action recognition. In 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014 (pp. 634-639). [6836043] (2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014). IEEE Computer Society. https://doi.org/10.1109/WACV.2014.6836043

@inproceedings{00befecc5482486bb848f6e49c20ad96,

title = "Mining discriminative 3D Poselet for cross-view action recognition",

abstract = "This paper presents a novel approach to cross-view action recognition. Traditional cross-view action recognition methods typically rely on local appearance/motion features. In this paper, we take advantage of the recent developments of depth cameras to build a more discriminative cross-view action representation. In this representation, an action is characterized by the spatio-temporal configuration of 3D Poselets, which are discriminatively discovered with a novel Poselet mining algorithm and can be detected with view-invariant 3D Poselet detectors. The Kinect skeleton is employed to facilitate the 3D Poselet mining and 3D Poselet detectors learning, but the recognition is solely based on 2D video input. Extensive experiments have demonstrated that this new action representation significantly improves the accuracy and robustness for cross-view action recognition.",

author = "Jiang Wang and Xiaohan Nie and Yin Xia and Ying Wu",

year = "2014",

month = jan,

day = "1",

doi = "10.1109/WACV.2014.6836043",

language = "English (US)",

isbn = "9781479949854",

series = "2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014",

publisher = "IEEE Computer Society",

pages = "634--639",

booktitle = "2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014",

address = "United States",

note = "2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014 ; Conference date: 24-03-2014 Through 26-03-2014",

}

Wang, J, Nie, X, Xia, Y & Wu, Y 2014, Mining discriminative 3D Poselet for cross-view action recognition. in 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014., 6836043, 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014, IEEE Computer Society, pp. 634-639, 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014, Steamboat Springs, CO, United States, 3/24/14. https://doi.org/10.1109/WACV.2014.6836043

Mining discriminative 3D Poselet for cross-view action recognition. / Wang, Jiang; Nie, Xiaohan; Xia, Yin; Wu, Ying.

2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014. IEEE Computer Society, 2014. p. 634-639 6836043 (2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014).

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

TY - GEN

T1 - Mining discriminative 3D Poselet for cross-view action recognition

AU - Wang, Jiang

AU - Nie, Xiaohan

AU - Xia, Yin

AU - Wu, Ying

PY - 2014/1/1

Y1 - 2014/1/1

N2 - This paper presents a novel approach to cross-view action recognition. Traditional cross-view action recognition methods typically rely on local appearance/motion features. In this paper, we take advantage of the recent developments of depth cameras to build a more discriminative cross-view action representation. In this representation, an action is characterized by the spatio-temporal configuration of 3D Poselets, which are discriminatively discovered with a novel Poselet mining algorithm and can be detected with view-invariant 3D Poselet detectors. The Kinect skeleton is employed to facilitate the 3D Poselet mining and 3D Poselet detectors learning, but the recognition is solely based on 2D video input. Extensive experiments have demonstrated that this new action representation significantly improves the accuracy and robustness for cross-view action recognition.

AB - This paper presents a novel approach to cross-view action recognition. Traditional cross-view action recognition methods typically rely on local appearance/motion features. In this paper, we take advantage of the recent developments of depth cameras to build a more discriminative cross-view action representation. In this representation, an action is characterized by the spatio-temporal configuration of 3D Poselets, which are discriminatively discovered with a novel Poselet mining algorithm and can be detected with view-invariant 3D Poselet detectors. The Kinect skeleton is employed to facilitate the 3D Poselet mining and 3D Poselet detectors learning, but the recognition is solely based on 2D video input. Extensive experiments have demonstrated that this new action representation significantly improves the accuracy and robustness for cross-view action recognition.

UR - http://www.scopus.com/inward/record.url?scp=84904627571&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84904627571&partnerID=8YFLogxK

U2 - 10.1109/WACV.2014.6836043

DO - 10.1109/WACV.2014.6836043

M3 - Conference contribution

AN - SCOPUS:84904627571

SN - 9781479949854

T3 - 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014

SP - 634

EP - 639

BT - 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014

PB - IEEE Computer Society

T2 - 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014

Y2 - 24 March 2014 through 26 March 2014

ER -