Action recognition using three-way cross-correlations feature of local motion attributes

Tetsu Matsukawa, Takio Kurita

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

6 Citations (Scopus)

Abstract

This paper proposes a spatio-temporal feature using three-way cross-correlations of local motion attributes for action recognition. Recently, the cubic higher-order local auto-correlations (CHLAC) feature has been shown high classification performances for action recognition. In previous researches, CHLAC feature was applied to binary motion image sequences that indicates moving or static points. However, each binary motion image lost informations about the type of motion such as timing of change or motion direction. Therefore, we can improve the classification accuracy further by extending CHLAC to multivalued motion image sequences that considered several types of local motion attributes. The proposed method is also viewed as an extension of popular bag-of-features approach. Experimental results using two datasets shows proposed method outperformed CHLAC features and bag-of-features approach.

Original languageEnglish
Title of host publicationProceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010
Pages1731-1734
Number of pages4
DOIs
Publication statusPublished - Nov 18 2010
Event2010 20th International Conference on Pattern Recognition, ICPR 2010 - Istanbul, Turkey
Duration: Aug 23 2010Aug 26 2010

Publication series

NameProceedings - International Conference on Pattern Recognition
ISSN (Print)1051-4651

Other

Other2010 20th International Conference on Pattern Recognition, ICPR 2010
CountryTurkey
CityIstanbul
Period8/23/108/26/10

Fingerprint

Autocorrelation

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

Matsukawa, T., & Kurita, T. (2010). Action recognition using three-way cross-correlations feature of local motion attributes. In Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010 (pp. 1731-1734). [5597474] (Proceedings - International Conference on Pattern Recognition). https://doi.org/10.1109/ICPR.2010.428

Action recognition using three-way cross-correlations feature of local motion attributes. / Matsukawa, Tetsu; Kurita, Takio.

Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010. 2010. p. 1731-1734 5597474 (Proceedings - International Conference on Pattern Recognition).

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

Matsukawa, T & Kurita, T 2010, Action recognition using three-way cross-correlations feature of local motion attributes. in Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010., 5597474, Proceedings - International Conference on Pattern Recognition, pp. 1731-1734, 2010 20th International Conference on Pattern Recognition, ICPR 2010, Istanbul, Turkey, 8/23/10. https://doi.org/10.1109/ICPR.2010.428
Matsukawa T, Kurita T. Action recognition using three-way cross-correlations feature of local motion attributes. In Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010. 2010. p. 1731-1734. 5597474. (Proceedings - International Conference on Pattern Recognition). https://doi.org/10.1109/ICPR.2010.428
Matsukawa, Tetsu ; Kurita, Takio. / Action recognition using three-way cross-correlations feature of local motion attributes. Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010. 2010. pp. 1731-1734 (Proceedings - International Conference on Pattern Recognition).
@inproceedings{e1749e88b437446999a5a19c7622831d,
title = "Action recognition using three-way cross-correlations feature of local motion attributes",
abstract = "This paper proposes a spatio-temporal feature using three-way cross-correlations of local motion attributes for action recognition. Recently, the cubic higher-order local auto-correlations (CHLAC) feature has been shown high classification performances for action recognition. In previous researches, CHLAC feature was applied to binary motion image sequences that indicates moving or static points. However, each binary motion image lost informations about the type of motion such as timing of change or motion direction. Therefore, we can improve the classification accuracy further by extending CHLAC to multivalued motion image sequences that considered several types of local motion attributes. The proposed method is also viewed as an extension of popular bag-of-features approach. Experimental results using two datasets shows proposed method outperformed CHLAC features and bag-of-features approach.",
author = "Tetsu Matsukawa and Takio Kurita",
year = "2010",
month = "11",
day = "18",
doi = "10.1109/ICPR.2010.428",
language = "English",
isbn = "9780769541099",
series = "Proceedings - International Conference on Pattern Recognition",
pages = "1731--1734",
booktitle = "Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010",

}

TY - GEN

T1 - Action recognition using three-way cross-correlations feature of local motion attributes

AU - Matsukawa, Tetsu

AU - Kurita, Takio

PY - 2010/11/18

Y1 - 2010/11/18

N2 - This paper proposes a spatio-temporal feature using three-way cross-correlations of local motion attributes for action recognition. Recently, the cubic higher-order local auto-correlations (CHLAC) feature has been shown high classification performances for action recognition. In previous researches, CHLAC feature was applied to binary motion image sequences that indicates moving or static points. However, each binary motion image lost informations about the type of motion such as timing of change or motion direction. Therefore, we can improve the classification accuracy further by extending CHLAC to multivalued motion image sequences that considered several types of local motion attributes. The proposed method is also viewed as an extension of popular bag-of-features approach. Experimental results using two datasets shows proposed method outperformed CHLAC features and bag-of-features approach.

AB - This paper proposes a spatio-temporal feature using three-way cross-correlations of local motion attributes for action recognition. Recently, the cubic higher-order local auto-correlations (CHLAC) feature has been shown high classification performances for action recognition. In previous researches, CHLAC feature was applied to binary motion image sequences that indicates moving or static points. However, each binary motion image lost informations about the type of motion such as timing of change or motion direction. Therefore, we can improve the classification accuracy further by extending CHLAC to multivalued motion image sequences that considered several types of local motion attributes. The proposed method is also viewed as an extension of popular bag-of-features approach. Experimental results using two datasets shows proposed method outperformed CHLAC features and bag-of-features approach.

UR - http://www.scopus.com/inward/record.url?scp=78149491736&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78149491736&partnerID=8YFLogxK

U2 - 10.1109/ICPR.2010.428

DO - 10.1109/ICPR.2010.428

M3 - Conference contribution

AN - SCOPUS:78149491736

SN - 9780769541099

T3 - Proceedings - International Conference on Pattern Recognition

SP - 1731

EP - 1734

BT - Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010

ER -