Video based human activities recognition using deep learning

Roubleh, A. A. and Khalifa, Othman Omran (2020) Video based human activities recognition using deep learning. In: 7th International Conference on Electronic Devices, Systems and Applications (ICEDSA2020), 28th - 29th March 2020, Shah Alam, Selangor.

	PDF (AIP Proceeding) - Published Version Restricted to Repository staff only Download (664kB) | Request a copy
	PDF (Acceptance letter) - Supplemental Material Restricted to Repository staff only Download (16kB) | Request a copy
	PDF (Certificate) - Supplemental Material Restricted to Repository staff only Download (377kB) | Request a copy

Abstract

Human activities recognition from motion capture data is a challenging problem in the computer vision due to the fact that, in various human activities, different body components have distinctive characteristics in terms of the moving pattern. In this paper, a learning method of detecting an activities from different angles based on various sources of information is proposed. with high accuracy. The bottomup approach is used in OpenPose which is the tool used in this paper’s experiments The proposed method achieve promising results on the MHAD datasets at 98% accuracy.

Item Type:	Conference or Workshop Item (Plenary Papers)
Additional Information:	4119/82391
Uncontrolled Keywords:	Learning method, deep learning, human activity recognition (HAR)
Subjects:	T Technology > T Technology (General)
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Engineering Kulliyyah of Engineering > Department of Electrical and Computer Engineering
Depositing User:	Prof. Dr Othman O. Khalifa
Date Deposited:	27 Aug 2020 15:10
Last Modified:	30 Dec 2020 16:15
URI:	http://irep.iium.edu.my/id/eprint/82391

Actions (login required)

View Item