Modality Consistency-Guided Contrastive Learning for Wearable-Based Human Activity Recognition
文献类型：期刊论文
| 作者 | Guo, Changru1; Zhang, Yingwei2,3; Chen, Yiqiang3; Xu, Chenyang4; Wang, Zhong1 |
| 刊名 | IEEE INTERNET OF THINGS JOURNAL
 |
| 出版日期 | 2024-06-15 |
| 卷号 | 11期号:12页码:21750-21762 |
| 关键词 | Human activity recognition Self-supervised learning Task analysis Data models Time series analysis Internet of Things Face recognition Contrastive learning (CL) human activity recognition (HAR) intermodality intramodality self-supervised |
| ISSN号 | 2327-4662 |
| DOI | 10.1109/JIOT.2024.3379019 |
| 英文摘要 | In wearable sensor-based human activity recognition (HAR) research, some factors limit the development of generalized models, such as the time and resource consuming, to acquire abundant annotated data, and the interdata set inconsistency of activity category. In this article, we take advantage of the complementarity and redundancy between different wearable modalities (e.g., accelerometers, gyroscopes, and magnetometers), and propose a modality consistency-guided contrastive learning (ModCL) method, which can construct a generalized model using annotation-free self-supervised learning and realize personalized domain adaptation with small amount annotation data. Specifically, ModCL exploits both intramodality and intermodality consistency of the wearable device data to construct contrastive learning tasks, encouraging the recognition model to recognize similar patterns and distinguish dissimilar ones. By leveraging these mixed constraint strategies, ModCL can learn the inherent activity patterns and extract meaningful generalized features across different data sets. To verify the effectiveness of ModCL method, we conduct experiments on five benchmark data sets (i.e., OPPORTUNITY and PAMAP2 as pretraining data sets, while UniMiB-SHAR, UCI-HAR, and WISDM as independent validation data sets). Experimental results show that ModCL achieves significant improvements in recognition accuracy compared with other state-of-the-art methods. |
| WOS研究方向 | Computer Science ; Engineering ; Telecommunications |
| 语种 | 英语 |
| WOS记录号 | WOS:001242362600070 |
| 出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
| 源URL | [http://119.78.100.204/handle/2XEOYT63/39904]  |
| 专题 | 中国科学院计算技术研究所期刊论文_英文 |
| 通讯作者 | Zhang, Yingwei; Wang, Zhong |
| 作者单位 | 1.Lanzhou Univ, Sch Comp Sci & Engn, Lanzhou 730000, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Beijing Key Lab Mobile Comp & Pervas Device, Beijing 100190, Peoples R China 3.Univ Chinese Acad Sci, Beijing 100190, Peoples R China 4.Tianjin Univ, Sch Comp Sci, Tianjin 300072, Peoples R China |
| 推荐引用方式 GB/T 7714 | Guo, Changru,Zhang, Yingwei,Chen, Yiqiang,et al. Modality Consistency-Guided Contrastive Learning for Wearable-Based Human Activity Recognition[J]. IEEE INTERNET OF THINGS JOURNAL,2024,11(12):21750-21762. |
| APA | Guo, Changru,Zhang, Yingwei,Chen, Yiqiang,Xu, Chenyang,&Wang, Zhong.(2024).Modality Consistency-Guided Contrastive Learning for Wearable-Based Human Activity Recognition.IEEE INTERNET OF THINGS JOURNAL,11(12),21750-21762. |
| MLA | Guo, Changru,et al."Modality Consistency-Guided Contrastive Learning for Wearable-Based Human Activity Recognition".IEEE INTERNET OF THINGS JOURNAL 11.12(2024):21750-21762. |
入库方式： OAI收割
来源：计算技术研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。