TY - GEN
T1 - Activity Recognition Based on FR-CNN and Attention-Based LSTM Network
AU - Tan, Tan Hsu
AU - Huang, Ching Jung
AU - Gochoo, Munkhjargal
AU - Chen, Yung Fu
N1 - Funding Information:
ACKNOWLEDGMENT This work was supported in part by the Ministry of Science and Technology of the Republic of China (Taiwan) under the Contract MOST 109-2221-E-027-97.
Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - A human activity recognition (HAR) based on the Faster Region-based Convolutional Neural Network (FR-CNN) and attention-based LSTM networks is proposed in this paper. A new structure of posture vector is generated by extracting skeleton joints of human movement using the pre-trained FR-CNN model. The Cornell Activity Dataset (CAD-60) is employed in the training and test phases. An attention-based bidirectional LSTM (Bi-LSTM) network is presented for activity classification. Experimental result shows that the attention-based Bi-LSTM network achieves the precision and recall rate of 97.02% and 96.83%, respectively, in recognizing twelve activities. The result is superior to the other existing systems, indicating the application potential of our work.
AB - A human activity recognition (HAR) based on the Faster Region-based Convolutional Neural Network (FR-CNN) and attention-based LSTM networks is proposed in this paper. A new structure of posture vector is generated by extracting skeleton joints of human movement using the pre-trained FR-CNN model. The Cornell Activity Dataset (CAD-60) is employed in the training and test phases. An attention-based bidirectional LSTM (Bi-LSTM) network is presented for activity classification. Experimental result shows that the attention-based Bi-LSTM network achieves the precision and recall rate of 97.02% and 96.83%, respectively, in recognizing twelve activities. The result is superior to the other existing systems, indicating the application potential of our work.
KW - attention-based bidirectional LSTM
KW - CAD-60
KW - deep learning
KW - FR-CNN
KW - human activity recognition
UR - http://www.scopus.com/inward/record.url?scp=85123449929&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85123449929&partnerID=8YFLogxK
U2 - 10.1109/WOCC53213.2021.9603203
DO - 10.1109/WOCC53213.2021.9603203
M3 - Conference contribution
AN - SCOPUS:85123449929
T3 - 2021 30th Wireless and Optical Communications Conference, WOCC 2021
SP - 146
EP - 149
BT - 2021 30th Wireless and Optical Communications Conference, WOCC 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 30th Wireless and Optical Communications Conference, WOCC 2021
Y2 - 7 October 2021 through 8 October 2021
ER -