Real-time human action recognition using stacked sparse autoencoders

Objectives: In this paper, an automated real-time human and human-action detection system is developed using Histogram of Oriented Gradients (HOG) and Stacked Sparse Auto-encoders respectively. Methods: For human detection, a feature descriptor is trained using SVM classifier and then is used for i...

Full description

Bibliographic Details
Main Authors:	Farooq, Adnan, Mohammad, Emad U Din, Ahmad Zarir, Abdullah, Ismail, Amelia Ritahani, Sulaiman, Suriani
Format:	Article
Language:	English
Published:	Indian Society for Education and Environment & Informatics Publishing Limited 2018
Subjects:	QA75 Electronic computers. Computer science
Online Access:	http://irep.iium.edu.my/62341/ http://irep.iium.edu.my/62341/ http://irep.iium.edu.my/62341/1/Real-Time%20Human%20Action%20Recognition.pdf

id	iium-62341
recordtype	eprints
spelling	iium-623412018-10-29T07:31:27Z http://irep.iium.edu.my/62341/ Real-time human action recognition using stacked sparse autoencoders Farooq, Adnan Mohammad, Emad U Din Ahmad Zarir, Abdullah Ismail, Amelia Ritahani Sulaiman, Suriani QA75 Electronic computers. Computer science Objectives: In this paper, an automated real-time human and human-action detection system is developed using Histogram of Oriented Gradients (HOG) and Stacked Sparse Auto-encoders respectively. Methods: For human detection, a feature descriptor is trained using SVM classifier and then is used for identification of humans in the frames. Stacked Sparse autoencoders are a category of deep neural networks, and in the proposed work is used for the feature extraction of human actions from the human action video dataset. The extracted features represent a dictionary which is used to map the input and produce a linear combination, following that soft-max classification is applied to train the model. To reduce the computational complexity, input frames has been changed into binary temporal difference images and fed to the neural network. Analysis: The proposed model matched the other state of the art models applied for human-action recognition classification problems. Applications: The study reveals that using multiple layers can improve the classification performance: 75% with two-layers and 83% with three-layers model. Indian Society for Education and Environment & Informatics Publishing Limited 2018-01 Article PeerReviewed application/pdf en http://irep.iium.edu.my/62341/1/Real-Time%20Human%20Action%20Recognition.pdf Farooq, Adnan and Mohammad, Emad U Din and Ahmad Zarir, Abdullah and Ismail, Amelia Ritahani and Sulaiman, Suriani (2018) Real-time human action recognition using stacked sparse autoencoders. Indian Journal of Science and Technology, 11 (4). pp. 1-6. ISSN 0974-6846 E-ISSN 0974-5645 http://www.indjst.org/index.php/indjst/article/view/121090/83462
repository_type	Digital Repository
institution_category	Local University
institution	International Islamic University Malaysia
building	IIUM Repository
collection	Online Access
language	English
topic	QA75 Electronic computers. Computer science
spellingShingle	QA75 Electronic computers. Computer science Farooq, Adnan Mohammad, Emad U Din Ahmad Zarir, Abdullah Ismail, Amelia Ritahani Sulaiman, Suriani Real-time human action recognition using stacked sparse autoencoders
description	Objectives: In this paper, an automated real-time human and human-action detection system is developed using Histogram of Oriented Gradients (HOG) and Stacked Sparse Auto-encoders respectively. Methods: For human detection, a feature descriptor is trained using SVM classifier and then is used for identification of humans in the frames. Stacked Sparse autoencoders are a category of deep neural networks, and in the proposed work is used for the feature extraction of human actions from the human action video dataset. The extracted features represent a dictionary which is used to map the input and produce a linear combination, following that soft-max classification is applied to train the model. To reduce the computational complexity, input frames has been changed into binary temporal difference images and fed to the neural network. Analysis: The proposed model matched the other state of the art models applied for human-action recognition classification problems. Applications: The study reveals that using multiple layers can improve the classification performance: 75% with two-layers and 83% with three-layers model.
format	Article
author	Farooq, Adnan Mohammad, Emad U Din Ahmad Zarir, Abdullah Ismail, Amelia Ritahani Sulaiman, Suriani
author_facet	Farooq, Adnan Mohammad, Emad U Din Ahmad Zarir, Abdullah Ismail, Amelia Ritahani Sulaiman, Suriani
author_sort	Farooq, Adnan
title	Real-time human action recognition using stacked sparse autoencoders
title_short	Real-time human action recognition using stacked sparse autoencoders
title_full	Real-time human action recognition using stacked sparse autoencoders
title_fullStr	Real-time human action recognition using stacked sparse autoencoders
title_full_unstemmed	Real-time human action recognition using stacked sparse autoencoders
title_sort	real-time human action recognition using stacked sparse autoencoders
publisher	Indian Society for Education and Environment & Informatics Publishing Limited
publishDate	2018
url	http://irep.iium.edu.my/62341/ http://irep.iium.edu.my/62341/ http://irep.iium.edu.my/62341/1/Real-Time%20Human%20Action%20Recognition.pdf
first_indexed	2023-09-18T21:28:22Z
last_indexed	2023-09-18T21:28:22Z
_version_	1777412344388780032

Real-time human action recognition using stacked sparse autoencoders

Similar Items