Mhad: Multi-human action dataset

Omar Elharrouss; Noor Almaadeed; Somaya Al-Maadeed

doi:10.1007/978-981-15-0637-6_28

Mhad: Multi-human action dataset

Omar Elharrouss, Noor Almaadeed, Somaya Al-Maadeed

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

6 Citations (Scopus)

Abstract

This paper presents a framework for a multi-action recognition method. In this framework, we introduce a new approach to detect and recognize the action of several persons within one scene. Also, considering the scarcity of related data, we provide a new data set involving many persons performing different actions in the same video. Our multi-action recognition method is based on a three-dimensional convolution neural network, and it involves a preprocessing phase to prepare the data to be recognized using the 3DCNN model. The new representation of data consists in extracting each person’s sequence during its presence in the scene. Then, we analyze each sequence to detect the actions in it. The experimental results proved to be accurate, efficient, and robust in real-time multi-human action recognition.

Original language	English
Title of host publication	4th International Congress on Information and Communication Technology, ICICT 2019, Volume 1
Editors	Xin-She Yang, Simon Sherratt, Nilanjan Dey, Amit Joshi
Publisher	Springer
Pages	333-341
Number of pages	9
ISBN (Print)	9789811506369
DOIs	https://doi.org/10.1007/978-981-15-0637-6_28
Publication status	Published - 2020
Externally published	Yes
Event	4th International Congress on Information and Communication Technology, ICICT 2019 - London, United Kingdom Duration: Feb 27 2019 → Feb 28 2019

Publication series

Name	Advances in Intelligent Systems and Computing
Volume	1041
ISSN (Print)	2194-5357
ISSN (Electronic)	2194-5365

Conference

Conference	4th International Congress on Information and Communication Technology, ICICT 2019
Country/Territory	United Kingdom
City	London
Period	2/27/19 → 2/28/19

Keywords

Convolutional neural network (CNN)
Human action recognition
Multi-human action recognition
Video surveillance

ASJC Scopus subject areas

Control and Systems Engineering
General Computer Science

Access to Document

10.1007/978-981-15-0637-6_28

Cite this

Mhad: Multi-human action dataset. / Elharrouss, Omar; Almaadeed, Noor; Al-Maadeed, Somaya.
4th International Congress on Information and Communication Technology, ICICT 2019, Volume 1. ed. / Xin-She Yang; Simon Sherratt; Nilanjan Dey; Amit Joshi. Springer, 2020. p. 333-341 (Advances in Intelligent Systems and Computing; Vol. 1041).

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

Elharrouss, O, Almaadeed, N & Al-Maadeed, S 2020, Mhad: Multi-human action dataset. in X-S Yang, S Sherratt, N Dey & A Joshi (eds), 4th International Congress on Information and Communication Technology, ICICT 2019, Volume 1. Advances in Intelligent Systems and Computing, vol. 1041, Springer, pp. 333-341, 4th International Congress on Information and Communication Technology, ICICT 2019, London, United Kingdom, 2/27/19. https://doi.org/10.1007/978-981-15-0637-6_28

@inproceedings{66eee812b195492c84785d92b140dbc4,

title = "Mhad: Multi-human action dataset",

abstract = "This paper presents a framework for a multi-action recognition method. In this framework, we introduce a new approach to detect and recognize the action of several persons within one scene. Also, considering the scarcity of related data, we provide a new data set involving many persons performing different actions in the same video. Our multi-action recognition method is based on a three-dimensional convolution neural network, and it involves a preprocessing phase to prepare the data to be recognized using the 3DCNN model. The new representation of data consists in extracting each person{\textquoteright}s sequence during its presence in the scene. Then, we analyze each sequence to detect the actions in it. The experimental results proved to be accurate, efficient, and robust in real-time multi-human action recognition.",

keywords = "Convolutional neural network (CNN), Human action recognition, Multi-human action recognition, Video surveillance",

author = "Omar Elharrouss and Noor Almaadeed and Somaya Al-Maadeed",

note = "Publisher Copyright: {\textcopyright} Springer Nature Singapore Pte Ltd. 2020.; 4th International Congress on Information and Communication Technology, ICICT 2019 ; Conference date: 27-02-2019 Through 28-02-2019",

year = "2020",

doi = "10.1007/978-981-15-0637-6_28",

language = "English",

isbn = "9789811506369",

series = "Advances in Intelligent Systems and Computing",

publisher = "Springer",

pages = "333--341",

editor = "Xin-She Yang and Simon Sherratt and Nilanjan Dey and Amit Joshi",

booktitle = "4th International Congress on Information and Communication Technology, ICICT 2019, Volume 1",

}

TY - GEN

T1 - Mhad

T2 - 4th International Congress on Information and Communication Technology, ICICT 2019

AU - Elharrouss, Omar

AU - Almaadeed, Noor

AU - Al-Maadeed, Somaya

N1 - Publisher Copyright: © Springer Nature Singapore Pte Ltd. 2020.

PY - 2020

Y1 - 2020

N2 - This paper presents a framework for a multi-action recognition method. In this framework, we introduce a new approach to detect and recognize the action of several persons within one scene. Also, considering the scarcity of related data, we provide a new data set involving many persons performing different actions in the same video. Our multi-action recognition method is based on a three-dimensional convolution neural network, and it involves a preprocessing phase to prepare the data to be recognized using the 3DCNN model. The new representation of data consists in extracting each person’s sequence during its presence in the scene. Then, we analyze each sequence to detect the actions in it. The experimental results proved to be accurate, efficient, and robust in real-time multi-human action recognition.

AB - This paper presents a framework for a multi-action recognition method. In this framework, we introduce a new approach to detect and recognize the action of several persons within one scene. Also, considering the scarcity of related data, we provide a new data set involving many persons performing different actions in the same video. Our multi-action recognition method is based on a three-dimensional convolution neural network, and it involves a preprocessing phase to prepare the data to be recognized using the 3DCNN model. The new representation of data consists in extracting each person’s sequence during its presence in the scene. Then, we analyze each sequence to detect the actions in it. The experimental results proved to be accurate, efficient, and robust in real-time multi-human action recognition.

KW - Convolutional neural network (CNN)

KW - Human action recognition

KW - Multi-human action recognition

KW - Video surveillance

UR - http://www.scopus.com/inward/record.url?scp=85077112304&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85077112304&partnerID=8YFLogxK

U2 - 10.1007/978-981-15-0637-6_28

DO - 10.1007/978-981-15-0637-6_28

M3 - Conference contribution

AN - SCOPUS:85077112304

SN - 9789811506369

T3 - Advances in Intelligent Systems and Computing

SP - 333

EP - 341

BT - 4th International Congress on Information and Communication Technology, ICICT 2019, Volume 1

A2 - Yang, Xin-She

A2 - Sherratt, Simon

A2 - Dey, Nilanjan

A2 - Joshi, Amit

PB - Springer

Y2 - 27 February 2019 through 28 February 2019

ER -

Mhad: Multi-human action dataset

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this