Appearance and motion information based human activity recognition

Mahmoud Al-Faris, John Chiverton, Linda Yang, David Ndzi

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

Abstract

Activity recognition is an essential objective of a smart building system which responds to what is happening in a scene. In this paper, a view invariant activity recognition system is proposed to recognise human actions. Selection of applicable features is made and solutions are proposed to deal with probable challenges including differing views on actions and directionality issues. This paper explores a number of features that can be utilised in action recognition systems and chooses suitable features to mitigate the challenges properly. Motion History Image (MHI) based on historical appearance information is used in combination with local motion vectors which are computed through each iteration sequence of the MHI information using an optical flow algorithm. A multiview dataset (MuHaVi) and a single view dataset (Weizmann) are used to demonstrate and validate the proposed method. Our method, can detect a wide range of actions in multi-view scenarios and shows competitive performance in comparison with state-of-the-art action classification techniques.
Original languageEnglish
Title of host publicationIET 3rd International Conference on ​​Intelligent Signal Processing (ISP 2017)
Place of PublicationLondon
PublisherIET
Number of pages6
ISBN (Electronic)978-1-78561-708-9
ISBN (Print)978-1-78561-707-2
DOIs
Publication statusPublished - 4 Dec 2017
Event​IET 3rd International Conference on ​​Intelligent Signal Processing - Savoy, London, United Kingdom
Duration: 4 Dec 20175 Dec 2017
https://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=8329306

Conference

Conference​IET 3rd International Conference on ​​Intelligent Signal Processing
Abbreviated titleISP 2017
CountryUnited Kingdom
CityLondon
Period4/12/175/12/17
Internet address

Fingerprint

Intelligent buildings
Optical flows

Keywords

  • motion history image
  • optical flow

Cite this

Al-Faris, M., Chiverton, J., Yang, L., & Ndzi, D. (2017). Appearance and motion information based human activity recognition. In IET 3rd International Conference on ​​Intelligent Signal Processing (ISP 2017) London: IET. https://doi.org/10.1049/cp.2017.0348
Al-Faris, Mahmoud ; Chiverton, John ; Yang, Linda ; Ndzi, David. / Appearance and motion information based human activity recognition. IET 3rd International Conference on ​​Intelligent Signal Processing (ISP 2017). London : IET, 2017.
@inproceedings{5c9d49476a2c4794992adba4300efad8,
title = "Appearance and motion information based human activity recognition",
abstract = "Activity recognition is an essential objective of a smart building system which responds to what is happening in a scene. In this paper, a view invariant activity recognition system is proposed to recognise human actions. Selection of applicable features is made and solutions are proposed to deal with probable challenges including differing views on actions and directionality issues. This paper explores a number of features that can be utilised in action recognition systems and chooses suitable features to mitigate the challenges properly. Motion History Image (MHI) based on historical appearance information is used in combination with local motion vectors which are computed through each iteration sequence of the MHI information using an optical flow algorithm. A multiview dataset (MuHaVi) and a single view dataset (Weizmann) are used to demonstrate and validate the proposed method. Our method, can detect a wide range of actions in multi-view scenarios and shows competitive performance in comparison with state-of-the-art action classification techniques.",
keywords = "motion history image, optical flow",
author = "Mahmoud Al-Faris and John Chiverton and Linda Yang and David Ndzi",
year = "2017",
month = "12",
day = "4",
doi = "10.1049/cp.2017.0348",
language = "English",
isbn = "978-1-78561-707-2",
booktitle = "IET 3rd International Conference on ​​Intelligent Signal Processing (ISP 2017)",
publisher = "IET",
address = "United Kingdom",

}

Al-Faris, M, Chiverton, J, Yang, L & Ndzi, D 2017, Appearance and motion information based human activity recognition. in IET 3rd International Conference on ​​Intelligent Signal Processing (ISP 2017). IET, London, ​IET 3rd International Conference on ​​Intelligent Signal Processing, London, United Kingdom, 4/12/17. https://doi.org/10.1049/cp.2017.0348

Appearance and motion information based human activity recognition. / Al-Faris, Mahmoud ; Chiverton, John; Yang, Linda; Ndzi, David.

IET 3rd International Conference on ​​Intelligent Signal Processing (ISP 2017). London : IET, 2017.

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

TY - GEN

T1 - Appearance and motion information based human activity recognition

AU - Al-Faris, Mahmoud

AU - Chiverton, John

AU - Yang, Linda

AU - Ndzi, David

PY - 2017/12/4

Y1 - 2017/12/4

N2 - Activity recognition is an essential objective of a smart building system which responds to what is happening in a scene. In this paper, a view invariant activity recognition system is proposed to recognise human actions. Selection of applicable features is made and solutions are proposed to deal with probable challenges including differing views on actions and directionality issues. This paper explores a number of features that can be utilised in action recognition systems and chooses suitable features to mitigate the challenges properly. Motion History Image (MHI) based on historical appearance information is used in combination with local motion vectors which are computed through each iteration sequence of the MHI information using an optical flow algorithm. A multiview dataset (MuHaVi) and a single view dataset (Weizmann) are used to demonstrate and validate the proposed method. Our method, can detect a wide range of actions in multi-view scenarios and shows competitive performance in comparison with state-of-the-art action classification techniques.

AB - Activity recognition is an essential objective of a smart building system which responds to what is happening in a scene. In this paper, a view invariant activity recognition system is proposed to recognise human actions. Selection of applicable features is made and solutions are proposed to deal with probable challenges including differing views on actions and directionality issues. This paper explores a number of features that can be utilised in action recognition systems and chooses suitable features to mitigate the challenges properly. Motion History Image (MHI) based on historical appearance information is used in combination with local motion vectors which are computed through each iteration sequence of the MHI information using an optical flow algorithm. A multiview dataset (MuHaVi) and a single view dataset (Weizmann) are used to demonstrate and validate the proposed method. Our method, can detect a wide range of actions in multi-view scenarios and shows competitive performance in comparison with state-of-the-art action classification techniques.

KW - motion history image

KW - optical flow

U2 - 10.1049/cp.2017.0348

DO - 10.1049/cp.2017.0348

M3 - Conference contribution

SN - 978-1-78561-707-2

BT - IET 3rd International Conference on ​​Intelligent Signal Processing (ISP 2017)

PB - IET

CY - London

ER -

Al-Faris M, Chiverton J, Yang L, Ndzi D. Appearance and motion information based human activity recognition. In IET 3rd International Conference on ​​Intelligent Signal Processing (ISP 2017). London: IET. 2017 https://doi.org/10.1049/cp.2017.0348