TY - JOUR
T1 - Passive Sensing for Class-Incremental Human Activity Recognition
AU - Ding, Xue
AU - Zhong, Yi
AU - Wu, Sheng
AU - Jiang, Chunxiao
AU - Xie, Weiliang
N1 - Publisher Copyright:
© 2004-2012 IEEE.
PY - 2023
Y1 - 2023
N2 - Passive sensing technology enables Wi-Fi-based human activity recognition (HAR), which has been widely noted in recent years. This letter presents a novel Wi-Fi-based class-incremental HAR system that allows for the gradual addition of new activity categories. To the best of our knowledge, this is the first attempt to recognize all previously learned activities under the constraint of limited samples for both the original and newly added activity classes. It is challenging in: 1) how to prevent catastrophic forgetting of old activities and 2) how to leverage as few samples as possible to accurately recognize new activities. Therefore, a phased training and update strategy is proposed to avoid the knowledge-forgetting issue. Furthermore, to alleviate the unsatisfactory performance problem caused by insufficient samples of new categories, we design an amplitude-phase enhanced convolution neural network (CNN), which integrates an attention mechanism and dual loss (DL) function to enhance the feature discrimination and the generalization capability of the model. Extensive experiments show that our system can operate with promising perceptual accuracy in different datasets.
AB - Passive sensing technology enables Wi-Fi-based human activity recognition (HAR), which has been widely noted in recent years. This letter presents a novel Wi-Fi-based class-incremental HAR system that allows for the gradual addition of new activity categories. To the best of our knowledge, this is the first attempt to recognize all previously learned activities under the constraint of limited samples for both the original and newly added activity classes. It is challenging in: 1) how to prevent catastrophic forgetting of old activities and 2) how to leverage as few samples as possible to accurately recognize new activities. Therefore, a phased training and update strategy is proposed to avoid the knowledge-forgetting issue. Furthermore, to alleviate the unsatisfactory performance problem caused by insufficient samples of new categories, we design an amplitude-phase enhanced convolution neural network (CNN), which integrates an attention mechanism and dual loss (DL) function to enhance the feature discrimination and the generalization capability of the model. Extensive experiments show that our system can operate with promising perceptual accuracy in different datasets.
KW - Class-incremental learning
KW - human activity recognition (HAR)
KW - loss function design
KW - passive sensing
UR - http://www.scopus.com/inward/record.url?scp=85160266768&partnerID=8YFLogxK
U2 - 10.1109/LGRS.2023.3278091
DO - 10.1109/LGRS.2023.3278091
M3 - Article
AN - SCOPUS:85160266768
SN - 1545-598X
VL - 20
JO - IEEE Geoscience and Remote Sensing Letters
JF - IEEE Geoscience and Remote Sensing Letters
M1 - 3504405
ER -