Revisiting LBP-based texture models for human action recognition

Thanh Phuong Nguyen; Antoine Manzanera; Ngoc Son Vu; Matthieu Garrigues

doi:10.1007/978-3-642-41827-3_36

Revisiting LBP-based texture models for human action recognition

Thanh Phuong Nguyen
, Antoine Manzanera
, Ngoc Son Vu
, Matthieu Garrigues

Université de Lyon

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

Abstract

A new method for action recognition is proposed by revisiting LBP-based dynamic texture operators. It captures the similarity of motion around keypoints tracked by a realtime semi-dense point tracking method. The use of self-similarity operator allows to highlight the geometric shape of rigid parts of foreground object in a video sequence. Inheriting from the efficient representation of LBP-based methods and the appearance invariance of patch matching method, the method is well designed for capturing action primitives in unconstrained videos. Action recognition experiments, made on several academic action datasets validate the interest of our approach.

Original language	English
Title of host publication	Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 18th Iberoamerican Congress, CIARP 2013, Proceedings
Pages	286-293
Number of pages	8
Edition	PART 2
DOIs	https://doi.org/10.1007/978-3-642-41827-3_36
Publication status	Published - 1 Dec 2013
Event	18th Iberoamerican Congress on Pattern Recognition, CIARP 2013 - Havana, Cuba Duration: 20 Nov 2013 → 23 Nov 2013

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 2
Volume	8259 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	18th Iberoamerican Congress on Pattern Recognition, CIARP 2013
Country/Territory	Cuba
City	Havana
Period	20/11/13 → 23/11/13

Keywords

Action recognition
Dynamic texture,. . .
Local binary pattern

Access to Document

10.1007/978-3-642-41827-3_36

Cite this

Nguyen, T. P., Manzanera, A., Vu, N. S., & Garrigues, M. (2013). Revisiting LBP-based texture models for human action recognition. In Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 18th Iberoamerican Congress, CIARP 2013, Proceedings (PART 2 ed., pp. 286-293). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8259 LNCS, No. PART 2). https://doi.org/10.1007/978-3-642-41827-3_36

Nguyen, Thanh Phuong ; Manzanera, Antoine ; Vu, Ngoc Son et al. / Revisiting LBP-based texture models for human action recognition. Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 18th Iberoamerican Congress, CIARP 2013, Proceedings. PART 2. ed. 2013. pp. 286-293 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2).

@inproceedings{faf64c4935f34da4a614ff085bed05e0,

title = "Revisiting LBP-based texture models for human action recognition",

abstract = "A new method for action recognition is proposed by revisiting LBP-based dynamic texture operators. It captures the similarity of motion around keypoints tracked by a realtime semi-dense point tracking method. The use of self-similarity operator allows to highlight the geometric shape of rigid parts of foreground object in a video sequence. Inheriting from the efficient representation of LBP-based methods and the appearance invariance of patch matching method, the method is well designed for capturing action primitives in unconstrained videos. Action recognition experiments, made on several academic action datasets validate the interest of our approach.",

keywords = "Action recognition, Dynamic texture,. . ., Local binary pattern",

author = "Nguyen, \{Thanh Phuong\} and Antoine Manzanera and Vu, \{Ngoc Son\} and Matthieu Garrigues",

year = "2013",

month = dec,

day = "1",

doi = "10.1007/978-3-642-41827-3\_36",

language = "English",

isbn = "9783642418266",

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

number = "PART 2",

pages = "286--293",

booktitle = "Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 18th Iberoamerican Congress, CIARP 2013, Proceedings",

edition = "PART 2",

note = "18th Iberoamerican Congress on Pattern Recognition, CIARP 2013 ; Conference date: 20-11-2013 Through 23-11-2013",

}

Nguyen, TP, Manzanera, A, Vu, NS & Garrigues, M 2013, Revisiting LBP-based texture models for human action recognition. in Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 18th Iberoamerican Congress, CIARP 2013, Proceedings. PART 2 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 2, vol. 8259 LNCS, pp. 286-293, 18th Iberoamerican Congress on Pattern Recognition, CIARP 2013, Havana, Cuba, 20/11/13. https://doi.org/10.1007/978-3-642-41827-3_36

Revisiting LBP-based texture models for human action recognition. / Nguyen, Thanh Phuong; Manzanera, Antoine; Vu, Ngoc Son et al.
Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 18th Iberoamerican Congress, CIARP 2013, Proceedings. PART 2. ed. 2013. p. 286-293 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8259 LNCS, No. PART 2).

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

TY - GEN

T1 - Revisiting LBP-based texture models for human action recognition

AU - Nguyen, Thanh Phuong

AU - Manzanera, Antoine

AU - Vu, Ngoc Son

AU - Garrigues, Matthieu

PY - 2013/12/1

Y1 - 2013/12/1

N2 - A new method for action recognition is proposed by revisiting LBP-based dynamic texture operators. It captures the similarity of motion around keypoints tracked by a realtime semi-dense point tracking method. The use of self-similarity operator allows to highlight the geometric shape of rigid parts of foreground object in a video sequence. Inheriting from the efficient representation of LBP-based methods and the appearance invariance of patch matching method, the method is well designed for capturing action primitives in unconstrained videos. Action recognition experiments, made on several academic action datasets validate the interest of our approach.

AB - A new method for action recognition is proposed by revisiting LBP-based dynamic texture operators. It captures the similarity of motion around keypoints tracked by a realtime semi-dense point tracking method. The use of self-similarity operator allows to highlight the geometric shape of rigid parts of foreground object in a video sequence. Inheriting from the efficient representation of LBP-based methods and the appearance invariance of patch matching method, the method is well designed for capturing action primitives in unconstrained videos. Action recognition experiments, made on several academic action datasets validate the interest of our approach.

KW - Action recognition

KW - Dynamic texture,. . .

KW - Local binary pattern

UR - https://www.scopus.com/pages/publications/84893192646

U2 - 10.1007/978-3-642-41827-3_36

DO - 10.1007/978-3-642-41827-3_36

M3 - Conference contribution

AN - SCOPUS:84893192646

SN - 9783642418266

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

SP - 286

EP - 293

BT - Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 18th Iberoamerican Congress, CIARP 2013, Proceedings

T2 - 18th Iberoamerican Congress on Pattern Recognition, CIARP 2013

Y2 - 20 November 2013 through 23 November 2013

ER -

Nguyen TP, Manzanera A, Vu NS, Garrigues M. Revisiting LBP-based texture models for human action recognition. In Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 18th Iberoamerican Congress, CIARP 2013, Proceedings. PART 2 ed. 2013. p. 286-293. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2). doi: 10.1007/978-3-642-41827-3_36

Revisiting LBP-based texture models for human action recognition

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this