Huatao Xu, Pengfei Zhou, R. Tan, Mo Li, Guobin Shen
{"title":"LIMU-BERT","authors":"Huatao Xu, Pengfei Zhou, R. Tan, Mo Li, Guobin Shen","doi":"10.1145/3568113.3568124","DOIUrl":null,"url":null,"abstract":"Deep learning greatly empowers Inertial Measurement Unit (IMU) sensors for a wide range of sensing applications. Most existing works require substantial amounts of wellcurated labeled data to train IMU-based sensing models, which incurs high annotation and training costs. Compared with labeled data, unlabeled IMU data are abundant and easily accessible. This article presents a novel representation learning model that can make use of unlabeled IMU data and extract generalized rather than task-specific features. With the representations learned via our model, task-specific models trained with limited labeled samples can achieve superior performances in typical IMU sensing applications, such as Human Activity Recognition (HAR).","PeriodicalId":29918,"journal":{"name":"GetMobile-Mobile Computing & Communications Review","volume":"10 1","pages":"39 - 42"},"PeriodicalIF":0.7000,"publicationDate":"2022-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"LIMU-BERT\",\"authors\":\"Huatao Xu, Pengfei Zhou, R. Tan, Mo Li, Guobin Shen\",\"doi\":\"10.1145/3568113.3568124\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Deep learning greatly empowers Inertial Measurement Unit (IMU) sensors for a wide range of sensing applications. Most existing works require substantial amounts of wellcurated labeled data to train IMU-based sensing models, which incurs high annotation and training costs. Compared with labeled data, unlabeled IMU data are abundant and easily accessible. This article presents a novel representation learning model that can make use of unlabeled IMU data and extract generalized rather than task-specific features. With the representations learned via our model, task-specific models trained with limited labeled samples can achieve superior performances in typical IMU sensing applications, such as Human Activity Recognition (HAR).\",\"PeriodicalId\":29918,\"journal\":{\"name\":\"GetMobile-Mobile Computing & Communications Review\",\"volume\":\"10 1\",\"pages\":\"39 - 42\"},\"PeriodicalIF\":0.7000,\"publicationDate\":\"2022-10-07\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"GetMobile-Mobile Computing & Communications Review\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3568113.3568124\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"TELECOMMUNICATIONS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"GetMobile-Mobile Computing & Communications Review","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3568113.3568124","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"TELECOMMUNICATIONS","Score":null,"Total":0}
Deep learning greatly empowers Inertial Measurement Unit (IMU) sensors for a wide range of sensing applications. Most existing works require substantial amounts of wellcurated labeled data to train IMU-based sensing models, which incurs high annotation and training costs. Compared with labeled data, unlabeled IMU data are abundant and easily accessible. This article presents a novel representation learning model that can make use of unlabeled IMU data and extract generalized rather than task-specific features. With the representations learned via our model, task-specific models trained with limited labeled samples can achieve superior performances in typical IMU sensing applications, such as Human Activity Recognition (HAR).