Action recognition based on feature extraction from time series

Author

Keceli, Ali Seydi ; Can, Ahmet Burak

Author_Institution

Bilgisayar Muhendisligi Bolumu, Hacettepe Univ., Ankara, Turkey

fYear

2014

fDate

23-25 April 2014

Firstpage

485

Lastpage

488

Abstract

Usage of 3^th dimension information obtained from depth sensors in human action recognition has gained importance in the recent years. In this study, basic human actions are tried to recognize on a human model derived from RGBD sensor. Joint angles and joint displacements used as time series and feature extraction from times series is applied to recognize actions. Actions are classified with the random forest and support vector machine approaches and classification accuracy is measured on MSRAction-3D and MSRC-12 datasets.

Keywords

feature extraction; image classification; image sensors; learning (artificial intelligence); object recognition; support vector machines; time series; MSRAction-3D dataset; MSRC-12 dataset; Microsoft Kinect; RGBD sensor; depth sensors; feature extraction; human action recognition; joint angles; joint displacements; random forest; support vector machine; time series; Conferences; Entropy; Histograms; Pattern recognition; Principal component analysis; Support vector machines; Three-dimensional displays; Action recognition; Microsoft Kinect; random fores; support vector machine; time series;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing and Communications Applications Conference (SIU), 2014 22nd

Conference_Location

Trabzon

Type

conf

DOI

10.1109/SIU.2014.6830271

Filename

6830271