loading
Papers Papers/2022 Papers Papers/2022

Research.Publish.Connect.

Paper

Paper Unlock
Efficient Multi-stream Temporal Learning and Post-fusion Strategy for 3D Skeleton-based Hand Activity Recognition

Topics: Deep Learning for Visual Understanding ; Egocentric Vision for Interaction Understanding; Event and Human Activity Recognition; Features Extraction; First Person Vision; Human and Computer Interaction; Transfer Learning; Understanding from Wearable and Mobile Cameras; Vision for Robotics

Authors: Yasser Boutaleb 1 ; 2 ; Catherine Soladie 1 ; Nam-Duong Duong 2 ; Amine Kacete 2 ; Jérôme Royan 2 and Renaud Seguier 1

Affiliations: 1 IETR/CentraleSupelec, Avenue de la Boulaie, 35510 Cesson-Sevigné, France ; 2 IRT b-com, 1219 Avenue des Champs Blancs, 35510 Cesson-Sevigné, France

Keyword(s): First-person Hand Activity, Multi-stream Learning, 3D Hand Skeleton, Hand-crafted Features, Temporal Learning.

Abstract: Recognizing first-person hand activity is a challenging task, especially when not enough data are available. In this paper, we tackle this challenge by proposing a new hybrid learning pipeline for skeleton-based hand activity recognition, which is composed of three blocks. First, for a given sequence of hand’s joint positions, the spatial features are extracted using a dedicated combination of local and global spacial hand-crafted features. Then, the temporal dependencies are learned using a multi-stream learning strategy. Finally, a hand activity sequence classifier is learned, via our Post-fusion strategy, applied to the previously learned temporal dependencies. The experiments, evaluated on two real-world data sets, show that our approach performs better than the state-of-the-art approaches. For more ablation study, we compared our Post-fusion strategy with three traditional fusion baselines and showed an improvement above 2.4% of accuracy.

CC BY-NC-ND 4.0

Sign In Guest: Register as new SciTePress user now for free.

Sign In SciTePress user: please login.

PDF ImageMy Papers

You are not signed in, therefore limits apply to your IP address 3.133.152.189

In the current month:
Recent papers: 100 available of 100 total
2+ years older papers: 200 available of 200 total

Paper citation in several formats:
Boutaleb, Y.; Soladie, C.; Duong, N.; Kacete, A.; Royan, J. and Seguier, R. (2021). Efficient Multi-stream Temporal Learning and Post-fusion Strategy for 3D Skeleton-based Hand Activity Recognition. In Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 4: VISAPP; ISBN 978-989-758-488-6; ISSN 2184-4321, SciTePress, pages 293-302. DOI: 10.5220/0010232702930302

@conference{visapp21,
author={Yasser Boutaleb. and Catherine Soladie. and Nam{-}Duong Duong. and Amine Kacete. and Jérôme Royan. and Renaud Seguier.},
title={Efficient Multi-stream Temporal Learning and Post-fusion Strategy for 3D Skeleton-based Hand Activity Recognition},
booktitle={Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 4: VISAPP},
year={2021},
pages={293-302},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010232702930302},
isbn={978-989-758-488-6},
issn={2184-4321},
}

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 4: VISAPP
TI - Efficient Multi-stream Temporal Learning and Post-fusion Strategy for 3D Skeleton-based Hand Activity Recognition
SN - 978-989-758-488-6
IS - 2184-4321
AU - Boutaleb, Y.
AU - Soladie, C.
AU - Duong, N.
AU - Kacete, A.
AU - Royan, J.
AU - Seguier, R.
PY - 2021
SP - 293
EP - 302
DO - 10.5220/0010232702930302
PB - SciTePress