Cross-Person Activity Recognition Method Using Snapshot Ensemble Learning

Siyuan Xu, Zhengran He, Wenjuan Shi, Yu Wang, Tomoaki Ohtsuki, Guan Guiy

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Human activity recognition (HAR) is one of the most promising technologies in the smart home, especially radio frequency (RF-based) method, which has the advantages of low cost, few privacy concerns and wide coverage. In recent years, deep learning (DL) has been introduced into HAR and these DL-based HAR methods usually have outstanding performance. However, as the recognition scenarios and target change, the model performance drops sharply. To solve this problem, we propose a generalized method for cross-person activity recognition (CPAR), which is called snapshot ensemble learning based an attention with bidirectional long short-term memory (SE-ABLSTM). Specifically, by defining the cosine annealing learning rate, the models with diversity are saved and integrated in the same training process. In addition, we provide a dataset for CPAR and simulation results show that our method improves generalization performance by 5% compared to the original method. The source code and dataset for all the experiments can be available at https://github.com/NJUPT-Sivan/Cross-person-HAR.

Original languageEnglish
Title of host publication2022 IEEE 96th Vehicular Technology Conference, VTC 2022-Fall 2022 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665454681
DOIs
Publication statusPublished - 2022
Event96th IEEE Vehicular Technology Conference, VTC 2022-Fall 2022 - London, United Kingdom
Duration: 2022 Sept 262022 Sept 29

Publication series

NameIEEE Vehicular Technology Conference
Volume2022-September
ISSN (Print)1550-2252

Conference

Conference96th IEEE Vehicular Technology Conference, VTC 2022-Fall 2022
Country/TerritoryUnited Kingdom
CityLondon
Period22/9/2622/9/29

Keywords

  • channel state information
  • generalization
  • Human activity recognition
  • snapshot ensemble.

ASJC Scopus subject areas

  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'Cross-Person Activity Recognition Method Using Snapshot Ensemble Learning'. Together they form a unique fingerprint.

Cite this