Human activity recognition based on ℜ transform

Ying Wang; Kaiqi Huang; Tieniu Tan

doi:10.1109/CVPR.2007.383505

Human activity recognition based on ℜ transform

Ying Wang^*, Kaiqi Huang, Tieniu Tan

^*Corresponding author for this work

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

154 Citations (Scopus)

Abstract

This paper addresses human activity recognition based on a new feature descriptor. For a binary human silhouette, an extended radon transform, ℜ transform, is employed to represent low-level features. The advantage of the ℜ transform lies in its low computational complexity and geometric invariance. Then a set of HMMs based on the extracted features are trained to recognize activities. Compared with other commonly-used feature descriptors, ℜ transform is robust to frame loss in video, disjoint silhouettes and holes in the shape, and thus achieves better performance in recognizing similar activities. Rich experiments have proved the efficiency of the proposed method.

Original language	English
Title of host publication	2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07
Place of Publication	Piscataway, N.J
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	1-8
Number of pages	8
ISBN (Print)	1424411807, 9781424411801, 1424411793
DOIs	https://doi.org/10.1109/CVPR.2007.383505
Publication status	Published - 2007
Externally published	Yes
Event	2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07 - Minneapolis, MN, United States Duration: 17 Jun 2007 → 22 Jun 2007

Other

Other	2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07
Country/Territory	United States
City	Minneapolis, MN
Period	17/06/07 → 22/06/07

Access to Document

10.1109/CVPR.2007.383505

Cite this

@inproceedings{68c69da4f47049df96c50d4034207d54,

title = "Human activity recognition based on ℜ transform",

abstract = "This paper addresses human activity recognition based on a new feature descriptor. For a binary human silhouette, an extended radon transform, ℜ transform, is employed to represent low-level features. The advantage of the ℜ transform lies in its low computational complexity and geometric invariance. Then a set of HMMs based on the extracted features are trained to recognize activities. Compared with other commonly-used feature descriptors, ℜ transform is robust to frame loss in video, disjoint silhouettes and holes in the shape, and thus achieves better performance in recognizing similar activities. Rich experiments have proved the efficiency of the proposed method.",

author = "Ying Wang and Kaiqi Huang and Tieniu Tan",

year = "2007",

doi = "10.1109/CVPR.2007.383505",

language = "English",

isbn = "1424411807",

pages = "1--8",

booktitle = "2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

note = "2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07 ; Conference date: 17-06-2007 Through 22-06-2007",

}

Wang, Y, Huang, K & Tan, T 2007, Human activity recognition based on ℜ transform. in 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07., 4270503, Institute of Electrical and Electronics Engineers (IEEE), Piscataway, N.J, pp. 1-8, 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07, Minneapolis, MN, United States, 17/06/07. https://doi.org/10.1109/CVPR.2007.383505

Human activity recognition based on ℜ transform. / Wang, Ying; Huang, Kaiqi; Tan, Tieniu.
2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07. Piscataway, N.J: Institute of Electrical and Electronics Engineers (IEEE), 2007. p. 1-8 4270503.

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

TY - GEN

T1 - Human activity recognition based on ℜ transform

AU - Wang, Ying

AU - Huang, Kaiqi

AU - Tan, Tieniu

PY - 2007

Y1 - 2007

N2 - This paper addresses human activity recognition based on a new feature descriptor. For a binary human silhouette, an extended radon transform, ℜ transform, is employed to represent low-level features. The advantage of the ℜ transform lies in its low computational complexity and geometric invariance. Then a set of HMMs based on the extracted features are trained to recognize activities. Compared with other commonly-used feature descriptors, ℜ transform is robust to frame loss in video, disjoint silhouettes and holes in the shape, and thus achieves better performance in recognizing similar activities. Rich experiments have proved the efficiency of the proposed method.

AB - This paper addresses human activity recognition based on a new feature descriptor. For a binary human silhouette, an extended radon transform, ℜ transform, is employed to represent low-level features. The advantage of the ℜ transform lies in its low computational complexity and geometric invariance. Then a set of HMMs based on the extracted features are trained to recognize activities. Compared with other commonly-used feature descriptors, ℜ transform is robust to frame loss in video, disjoint silhouettes and holes in the shape, and thus achieves better performance in recognizing similar activities. Rich experiments have proved the efficiency of the proposed method.

UR - https://www.scopus.com/pages/publications/34948884539

U2 - 10.1109/CVPR.2007.383505

DO - 10.1109/CVPR.2007.383505

M3 - Conference proceeding contribution

AN - SCOPUS:34948884539

SN - 1424411807

SN - 9781424411801

SN - 1424411793

SP - 1

EP - 8

BT - 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07

PB - Institute of Electrical and Electronics Engineers (IEEE)

CY - Piscataway, N.J

T2 - 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07

Y2 - 17 June 2007 through 22 June 2007

ER -

Human activity recognition based on ℜ transform

Abstract

Other

Access to Document

Other files and links

Fingerprint

Cite this