CovP3DJ: Skeleton-parts-based-covariance Descriptor for Human Action Recognition

Hany A. El-Ghaish, Amin Shoukry, Mohamed E. Hussein

2018

Abstract

A highly discriminative and computationally efficient descriptor is needed in many computer vision applications involving human action recognition. This paper proposes a hand-crafted skeleton-based descriptor for human action recognition. It is constructed from five fixed size covariance matrices calculated using strongly related joints coordinates over five body parts (spine, left/ right arms, and left/ right legs). Since covariance matrices are symmetric, the lower/ upper triangular parts of these matrices are concatenated to generate an efficient descriptor. It achieves a saving from 78.26 % to 80.35 % in storage space and from 75 % to 90 % in processing time (depending on the dataset) relative to techniques adopting a covariance descriptor based on all the skeleton joints. To show the effectiveness of the proposed method, its performance is evaluated on five public datasets: MSR-Action3D, MSRC-12 Kinect Gesture, UTKinect-Action, Florence3D-Action, and NTU RGB+D. The obtained recognition rates on all datasets outperform many existing methods and compete with the current state of the art techniques.

Paper Citation

in Harvard Style

El-Ghaish H., Shoukry A. and Hussein M. (2018). CovP3DJ: Skeleton-parts-based-covariance Descriptor for Human Action Recognition. In Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2018) - Volume 5: VISAPP; ISBN 978-989-758-290-5, SciTePress, pages 343-350. DOI: 10.5220/0006625703430350

in Bibtex Style

@conference{visapp18,
author={Hany A. El-Ghaish and Amin Shoukry and Mohamed E. Hussein},
title={CovP3DJ: Skeleton-parts-based-covariance Descriptor for Human Action Recognition},
booktitle={Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2018) - Volume 5: VISAPP},
year={2018},
pages={343-350},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006625703430350},
isbn={978-989-758-290-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2018) - Volume 5: VISAPP
TI - CovP3DJ: Skeleton-parts-based-covariance Descriptor for Human Action Recognition
SN - 978-989-758-290-5
AU - El-Ghaish H.
AU - Shoukry A.
AU - Hussein M.
PY - 2018
SP - 343
EP - 350
DO - 10.5220/0006625703430350
PB - SciTePress