Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

3D-HOG EMBEDDING FRAMEWORKS FOR SINGLE AND MULTI-VIEWPOINTS ACTION RECOGNITION BASED ON HUMAN SILHOUETTES

Abstract: 

Given the high demand for automated systems for human action recognition, great efforts have been undertaken in recent decades to progress the field. In this paper, we present frameworks for single and multi-viewpoints action recognition based on Space-Time Volume (STV) of human silhouettes and 3D-Histogram of Oriented Gradient (3D-HOG) embedding. We exploit fast-computational approaches involving Principal Component Analysis (PCA) over the local feature spaces for compactly describing actions as combinations of local gestures and L2-Regularized Logistic Regression (L2-RLR) for learning the action model from local features. Outperforming results on Weizmann and i3DPost datasets confirm efficacy of the proposed approaches as compared to the baseline method and other works, in terms of accuracy and robustness to appearance changes.

up
1 user has voted: zeyu fu

Paper Details

Authors:
Federico Angelini, Zeyu Fu, Sergio A. Velastin, Jonathon A. Chambers, Syed Mohsen Naqvi
Submitted On:
12 April 2018 - 4:29pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Federico Angelini
Paper Code:
2125
Document Year:
2018
Cite

Document Files

Poster

(107)

Subscribe

[1] Federico Angelini, Zeyu Fu, Sergio A. Velastin, Jonathon A. Chambers, Syed Mohsen Naqvi, "3D-HOG EMBEDDING FRAMEWORKS FOR SINGLE AND MULTI-VIEWPOINTS ACTION RECOGNITION BASED ON HUMAN SILHOUETTES", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2495. Accessed: Aug. 19, 2019.
@article{2495-18,
url = {http://sigport.org/2495},
author = {Federico Angelini; Zeyu Fu; Sergio A. Velastin; Jonathon A. Chambers; Syed Mohsen Naqvi },
publisher = {IEEE SigPort},
title = {3D-HOG EMBEDDING FRAMEWORKS FOR SINGLE AND MULTI-VIEWPOINTS ACTION RECOGNITION BASED ON HUMAN SILHOUETTES},
year = {2018} }
TY - EJOUR
T1 - 3D-HOG EMBEDDING FRAMEWORKS FOR SINGLE AND MULTI-VIEWPOINTS ACTION RECOGNITION BASED ON HUMAN SILHOUETTES
AU - Federico Angelini; Zeyu Fu; Sergio A. Velastin; Jonathon A. Chambers; Syed Mohsen Naqvi
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2495
ER -
Federico Angelini, Zeyu Fu, Sergio A. Velastin, Jonathon A. Chambers, Syed Mohsen Naqvi. (2018). 3D-HOG EMBEDDING FRAMEWORKS FOR SINGLE AND MULTI-VIEWPOINTS ACTION RECOGNITION BASED ON HUMAN SILHOUETTES. IEEE SigPort. http://sigport.org/2495
Federico Angelini, Zeyu Fu, Sergio A. Velastin, Jonathon A. Chambers, Syed Mohsen Naqvi, 2018. 3D-HOG EMBEDDING FRAMEWORKS FOR SINGLE AND MULTI-VIEWPOINTS ACTION RECOGNITION BASED ON HUMAN SILHOUETTES. Available at: http://sigport.org/2495.
Federico Angelini, Zeyu Fu, Sergio A. Velastin, Jonathon A. Chambers, Syed Mohsen Naqvi. (2018). "3D-HOG EMBEDDING FRAMEWORKS FOR SINGLE AND MULTI-VIEWPOINTS ACTION RECOGNITION BASED ON HUMAN SILHOUETTES." Web.
1. Federico Angelini, Zeyu Fu, Sergio A. Velastin, Jonathon A. Chambers, Syed Mohsen Naqvi. 3D-HOG EMBEDDING FRAMEWORKS FOR SINGLE AND MULTI-VIEWPOINTS ACTION RECOGNITION BASED ON HUMAN SILHOUETTES [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2495