Multi-view gymnastic activity recognition with fused HMM

Ying Wang; Kaiqi Huang; Tieniu Tan

doi:10.1007/978-3-540-76386-4_63

Multi-view gymnastic activity recognition with fused HMM

Ying Wang^*, Kaiqi Huang, Tieniu Tan

^*Corresponding author for this work

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

5 Citations (Scopus)

Abstract

More and more researchers focus their studies on multi-view activity recognition, because a fixed view could not provide enough information for recognition. In this paper, we use multi-view features to recognize six kinds of gymnastic activities. Firstly, shape-based features are extracted from two orthogonal cameras in the form of R transform. Then a multi-view approach based on Fused HMM is proposed to combine different features for similar gymnastic activity recognition. Compared with other activity models, our method achieves better performance even in the case of frame loss.

Original language	English
Title of host publication	Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings
Place of Publication	Tokyo, Japan
Publisher	Springer, Springer Nature
Pages	667-677
Number of pages	11
Volume	4843 LNCS
Edition	PART 1
ISBN (Electronic)	9783540763864
ISBN (Print)	9783540763857
DOIs	https://doi.org/10.1007/978-3-540-76386-4_63
Publication status	Published - 2007
Externally published	Yes
Event	8th Asian Conference on Computer Vision, ACCV 2007 - Tokyo, Japan Duration: 18 Nov 2007 → 22 Nov 2007

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 1
Volume	4843 LNCS
ISSN (Print)	03029743
ISSN (Electronic)	16113349

Other

Other	8th Asian Conference on Computer Vision, ACCV 2007
Country/Territory	Japan
City	Tokyo
Period	18/11/07 → 22/11/07

Access to Document

10.1007/978-3-540-76386-4_63

Cite this

Wang, Y., Huang, K., & Tan, T. (2007). Multi-view gymnastic activity recognition with fused HMM. In Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings (PART 1 ed., Vol. 4843 LNCS, pp. 667-677). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4843 LNCS, No. PART 1). Springer, Springer Nature. https://doi.org/10.1007/978-3-540-76386-4_63

Wang, Ying ; Huang, Kaiqi ; Tan, Tieniu. / Multi-view gymnastic activity recognition with fused HMM. Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings. Vol. 4843 LNCS PART 1. ed. Tokyo, Japan : Springer, Springer Nature, 2007. pp. 667-677 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1).

@inproceedings{310a9b4b0cfa461c9e8f6ae96f6b3f90,

title = "Multi-view gymnastic activity recognition with fused HMM",

abstract = "More and more researchers focus their studies on multi-view activity recognition, because a fixed view could not provide enough information for recognition. In this paper, we use multi-view features to recognize six kinds of gymnastic activities. Firstly, shape-based features are extracted from two orthogonal cameras in the form of R transform. Then a multi-view approach based on Fused HMM is proposed to combine different features for similar gymnastic activity recognition. Compared with other activity models, our method achieves better performance even in the case of frame loss.",

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

year = "2007",

doi = "10.1007/978-3-540-76386-4_63",

language = "English",

isbn = "9783540763857",

volume = "4843 LNCS",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer, Springer Nature",

number = "PART 1",

pages = "667--677",

booktitle = "Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings",

address = "United States",

edition = "PART 1",

note = "8th Asian Conference on Computer Vision, ACCV 2007 ; Conference date: 18-11-2007 Through 22-11-2007",

}

Wang, Y, Huang, K & Tan, T 2007, Multi-view gymnastic activity recognition with fused HMM. in Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings. PART 1 edn, vol. 4843 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 4843 LNCS, Springer, Springer Nature, Tokyo, Japan, pp. 667-677, 8th Asian Conference on Computer Vision, ACCV 2007, Tokyo, Japan, 18/11/07. https://doi.org/10.1007/978-3-540-76386-4_63

Multi-view gymnastic activity recognition with fused HMM. / Wang, Ying; Huang, Kaiqi; Tan, Tieniu.
Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings. Vol. 4843 LNCS PART 1. ed. Tokyo, Japan: Springer, Springer Nature, 2007. p. 667-677 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4843 LNCS, No. PART 1).

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

TY - GEN

T1 - Multi-view gymnastic activity recognition with fused HMM

AU - Wang, Ying

AU - Huang, Kaiqi

AU - Tan, Tieniu

PY - 2007

Y1 - 2007

N2 - More and more researchers focus their studies on multi-view activity recognition, because a fixed view could not provide enough information for recognition. In this paper, we use multi-view features to recognize six kinds of gymnastic activities. Firstly, shape-based features are extracted from two orthogonal cameras in the form of R transform. Then a multi-view approach based on Fused HMM is proposed to combine different features for similar gymnastic activity recognition. Compared with other activity models, our method achieves better performance even in the case of frame loss.

AB - More and more researchers focus their studies on multi-view activity recognition, because a fixed view could not provide enough information for recognition. In this paper, we use multi-view features to recognize six kinds of gymnastic activities. Firstly, shape-based features are extracted from two orthogonal cameras in the form of R transform. Then a multi-view approach based on Fused HMM is proposed to combine different features for similar gymnastic activity recognition. Compared with other activity models, our method achieves better performance even in the case of frame loss.

UR - http://www.scopus.com/inward/record.url?scp=38149029237&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-76386-4_63

DO - 10.1007/978-3-540-76386-4_63

M3 - Conference proceeding contribution

AN - SCOPUS:38149029237

SN - 9783540763857

VL - 4843 LNCS

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

SP - 667

EP - 677

BT - Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings

PB - Springer, Springer Nature

CY - Tokyo, Japan

T2 - 8th Asian Conference on Computer Vision, ACCV 2007

Y2 - 18 November 2007 through 22 November 2007

ER -

Wang Y, Huang K, Tan T. Multi-view gymnastic activity recognition with fused HMM. In Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings. PART 1 ed. Vol. 4843 LNCS. Tokyo, Japan: Springer, Springer Nature. 2007. p. 667-677. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). doi: 10.1007/978-3-540-76386-4_63

Multi-view gymnastic activity recognition with fused HMM

Abstract

Publication series

Other

Access to Document

Other files and links

Fingerprint

Cite this