Leveraging textural features for recognizing actions in low quality videos

Saimunur Rahman; John See; Chiung Ching Ho

doi:10.1007/978-981-10-1721-6_26

Leveraging textural features for recognizing actions in low quality videos

Saimunur Rahman^*, John See, Chiung Ching Ho

^*Corresponding author for this work

School of Mathematical & Computer Sciences

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

1 Citation (Scopus)

Abstract

Human action recognition is a well researched problem, which is considerably more challenging when video quality is poor. In this paper, we investigate human action recognition in low quality videos by leveraging the robustness of textural features to better characterize actions, instead of relying on shape and motion features may fail under noisy conditions. To accommodate videos, texture descriptors are extended to three orthogonal planes (TOP) to extract spatio-temporal features. Extensive experiments were conducted on lowquality versions of theKTH and HMDB51 datasets to evaluate the performance of our proposed approaches against standard baselines. Experimental results and further analysis demonstrated the usefulness of textural features in improving the capability of recognizing human actions from low quality videos.

Original language	English
Title of host publication	9th International Conference on Robotic, Vision, Signal Processing and Power Applications
Editors	Haidi Ibrahim, Shahid Iqbal, Soo Siang Teoh, Mohd Tafir Mustaffa
Publisher	Springer
Pages	237-245
Number of pages	9
ISBN (Electronic)	9789811017216
ISBN (Print)	9789811017193
DOIs	https://doi.org/10.1007/978-981-10-1721-6_26
Publication status	Published - 30 Sept 2016
Event	9th International Conference on Robotic, Vision, Signal Processing and Power Applications 2016 - , Malaysia Duration: 2 Feb 2016 → 3 Feb 2016

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	398
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	9th International Conference on Robotic, Vision, Signal Processing and Power Applications 2016
Abbreviated title	RoViSP 2016
Country/Territory	Malaysia
Period	2/02/16 → 3/02/16

ASJC Scopus subject areas

Industrial and Manufacturing Engineering

Access to Document

10.1007/978-981-10-1721-6_26

Cite this

Rahman, S., See, J., & Ho, C. C. (2016). Leveraging textural features for recognizing actions in low quality videos. In H. Ibrahim, S. Iqbal, S. S. Teoh, & M. T. Mustaffa (Eds.), 9th International Conference on Robotic, Vision, Signal Processing and Power Applications (pp. 237-245). (Lecture Notes in Electrical Engineering; Vol. 398). Springer. https://doi.org/10.1007/978-981-10-1721-6_26

@inproceedings{d88e40f1175e45309df3cd8e7b51be6f,

title = "Leveraging textural features for recognizing actions in low quality videos",

abstract = "Human action recognition is a well researched problem, which is considerably more challenging when video quality is poor. In this paper, we investigate human action recognition in low quality videos by leveraging the robustness of textural features to better characterize actions, instead of relying on shape and motion features may fail under noisy conditions. To accommodate videos, texture descriptors are extended to three orthogonal planes (TOP) to extract spatio-temporal features. Extensive experiments were conducted on lowquality versions of theKTH and HMDB51 datasets to evaluate the performance of our proposed approaches against standard baselines. Experimental results and further analysis demonstrated the usefulness of textural features in improving the capability of recognizing human actions from low quality videos.",

author = "Saimunur Rahman and John See and Ho, {Chiung Ching}",

note = "Publisher Copyright: {\textcopyright} Springer Science+Business Media Singapore 2017. Copyright: Copyright 2017 Elsevier B.V., All rights reserved.; 9th International Conference on Robotic, Vision, Signal Processing and Power Applications 2016, RoViSP 2016 ; Conference date: 02-02-2016 Through 03-02-2016",

year = "2016",

month = sep,

day = "30",

doi = "10.1007/978-981-10-1721-6_26",

language = "English",

isbn = "9789811017193",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer",

pages = "237--245",

editor = "Haidi Ibrahim and Shahid Iqbal and Teoh, {Soo Siang} and Mustaffa, {Mohd Tafir}",

booktitle = "9th International Conference on Robotic, Vision, Signal Processing and Power Applications",

}

Rahman, S, See, J & Ho, CC 2016, Leveraging textural features for recognizing actions in low quality videos. in H Ibrahim, S Iqbal, SS Teoh & MT Mustaffa (eds), 9th International Conference on Robotic, Vision, Signal Processing and Power Applications. Lecture Notes in Electrical Engineering, vol. 398, Springer, pp. 237-245, 9th International Conference on Robotic, Vision, Signal Processing and Power Applications 2016, Malaysia, 2/02/16. https://doi.org/10.1007/978-981-10-1721-6_26

Leveraging textural features for recognizing actions in low quality videos. / Rahman, Saimunur; See, John; Ho, Chiung Ching.
9th International Conference on Robotic, Vision, Signal Processing and Power Applications. ed. / Haidi Ibrahim; Shahid Iqbal; Soo Siang Teoh; Mohd Tafir Mustaffa. Springer, 2016. p. 237-245 (Lecture Notes in Electrical Engineering; Vol. 398).

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

TY - GEN

T1 - Leveraging textural features for recognizing actions in low quality videos

AU - Rahman, Saimunur

AU - See, John

AU - Ho, Chiung Ching

PY - 2016/9/30

Y1 - 2016/9/30

N2 - Human action recognition is a well researched problem, which is considerably more challenging when video quality is poor. In this paper, we investigate human action recognition in low quality videos by leveraging the robustness of textural features to better characterize actions, instead of relying on shape and motion features may fail under noisy conditions. To accommodate videos, texture descriptors are extended to three orthogonal planes (TOP) to extract spatio-temporal features. Extensive experiments were conducted on lowquality versions of theKTH and HMDB51 datasets to evaluate the performance of our proposed approaches against standard baselines. Experimental results and further analysis demonstrated the usefulness of textural features in improving the capability of recognizing human actions from low quality videos.

AB - Human action recognition is a well researched problem, which is considerably more challenging when video quality is poor. In this paper, we investigate human action recognition in low quality videos by leveraging the robustness of textural features to better characterize actions, instead of relying on shape and motion features may fail under noisy conditions. To accommodate videos, texture descriptors are extended to three orthogonal planes (TOP) to extract spatio-temporal features. Extensive experiments were conducted on lowquality versions of theKTH and HMDB51 datasets to evaluate the performance of our proposed approaches against standard baselines. Experimental results and further analysis demonstrated the usefulness of textural features in improving the capability of recognizing human actions from low quality videos.

UR - http://www.scopus.com/inward/record.url?scp=84992704046&partnerID=8YFLogxK

U2 - 10.1007/978-981-10-1721-6_26

DO - 10.1007/978-981-10-1721-6_26

M3 - Conference contribution

AN - SCOPUS:84992704046

SN - 9789811017193

T3 - Lecture Notes in Electrical Engineering

SP - 237

EP - 245

BT - 9th International Conference on Robotic, Vision, Signal Processing and Power Applications

A2 - Ibrahim, Haidi

A2 - Iqbal, Shahid

A2 - Teoh, Soo Siang

A2 - Mustaffa, Mohd Tafir

PB - Springer

T2 - 9th International Conference on Robotic, Vision, Signal Processing and Power Applications 2016

Y2 - 2 February 2016 through 3 February 2016

ER -

Leveraging textural features for recognizing actions in low quality videos

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this