TY - GEN
T1 - Trustworthy adaptation with few-shot learning for hand gesture recognition
AU - Rahimian, Elahe
AU - Zabihi, Soheil
AU - Asif, Amir
AU - Atashzar, S. Farokh
AU - Mohammadi, Arash
N1 - Funding Information:
This work is supported in part by the Canadian Department of National Defence through the AutoDefence project in the IDEaS program
Publisher Copyright:
© 2021 IEEE.
PY - 2021/8/11
Y1 - 2021/8/11
N2 - This work is motivated by potentials of Deep Neural Networks (DNNs)-based solutions in improving myoelectric control for trustworthy Human-Machine Interfacing (HMI). In this context, we propose the Trustworthy Few Shot-Hand Gesture Recognition (TFS-HGR) framework as a novel DNN-based architecture for performing Hand Gesture Recognition (HGR) via multi-channel surface Electromyography (sEMG) signals. The main objective of the TFS-HGR framework is to employ Few-Shot Learning (FSL) formulation with a focus on transferring information and knowledge between source and target domains (despite their inherit differences) to address limited availability of training data. The NinaPro DB5 dataset is used for evaluation purposes. The proposed TFS-HGR achieves a performance of 83.17% for new repetitions with few-shot observations, i.e., 5-way 10-shot classification. Moreover, the TFS-HGR with the accuracy of 75.29% also generalize to new gestures with few-shot observations, i.e., 5-way 10-shot classification.
AB - This work is motivated by potentials of Deep Neural Networks (DNNs)-based solutions in improving myoelectric control for trustworthy Human-Machine Interfacing (HMI). In this context, we propose the Trustworthy Few Shot-Hand Gesture Recognition (TFS-HGR) framework as a novel DNN-based architecture for performing Hand Gesture Recognition (HGR) via multi-channel surface Electromyography (sEMG) signals. The main objective of the TFS-HGR framework is to employ Few-Shot Learning (FSL) formulation with a focus on transferring information and knowledge between source and target domains (despite their inherit differences) to address limited availability of training data. The NinaPro DB5 dataset is used for evaluation purposes. The proposed TFS-HGR achieves a performance of 83.17% for new repetitions with few-shot observations, i.e., 5-way 10-shot classification. Moreover, the TFS-HGR with the accuracy of 75.29% also generalize to new gestures with few-shot observations, i.e., 5-way 10-shot classification.
KW - Attention Mechanism
KW - Few-Shot Learning (FSL)
KW - Hand Gesture Recognition (HGR)
KW - Surface Electromyographic (sEMG)
KW - Temporal Convolution
UR - http://www.scopus.com/inward/record.url?scp=85117478980&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85117478980&partnerID=8YFLogxK
U2 - 10.1109/ICAS49788.2021.9551144
DO - 10.1109/ICAS49788.2021.9551144
M3 - Conference contribution
AN - SCOPUS:85117478980
T3 - ICAS 2021 - 2021 IEEE International Conference on Autonomous Systems, Proceedings
BT - ICAS 2021 - 2021 IEEE International Conference on Autonomous Systems, Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 IEEE International Conference on Autonomous Systems, ICAS 2021
Y2 - 11 August 2021 through 13 August 2021
ER -