Enhanced Human Action Recognition Using Fusion of Skeletal Joint Dynamics and Structural Features

Muralikrishna, S N and Muniyal, Balachandra and Acharya, Dinesh U and Holla, Raghurama (2020) Enhanced Human Action Recognition Using Fusion of Skeletal Joint Dynamics and Structural Features. Journal of Robotics. ISSN 1687-9600

[img] PDF
9478.pdf - Published Version
Restricted to Registered users only

Download (3MB) | Request a copy

Abstract

In this research work, we propose a method for human action recognition based on the combination of structural and temporal features. ,e pose sequence in the video is considered to identify the action type. ,e structural variation features are obtained by detecting the angle made between the joints during the action, where the angle binning is performed using multiple thresholds. ,e displacement vector of joint locations is used to compute the temporal features. ,e structural variation features and the temporal variation features are fused using a neural network to perform action classification. We conducted the experiments on different categories of datasets, namely, KTH, UTKinect, and MSR Action3D datasets. ,e experimental results exhibit the superiority of the proposed method over some of the existing state-of-the-art techniques.

Item Type: Article
Uncontrolled Keywords: Human Action Recognition, SVM, Classification, skeleton, pose
Subjects: Engineering > MIT Manipal > Computer Science and Engineering
Engineering > MIT Manipal > Information and Communication Technology
Engineering > MIT Manipal > MCA
Depositing User: MIT Library
Date Deposited: 01 Jan 2021 04:16
Last Modified: 01 Jan 2021 04:16
URI: http://eprints.manipal.edu/id/eprint/156076

Actions (login required)

View Item View Item