Work

Practically Adopting Human Activity Recognition

Year: 2023

Type: conference-paper

Abstract: Existing inertial measurement unit (IMU) based human activity recognition (HAR) approaches still face a major challenge when adopted across users in practice. The severe heterogeneity in IMU data significantly undermines model generalizability in wild adoption. This paper presents UniHAR, a universal HAR framework for mobile devices. To address the challenge of data heterogeneity, we thoroughly study augmenting data with the physics of the IMU sensing process and present a novel adoption of data augmentations for exploiting both unlabeled and labeled data. We consider two application scenarios of UniHAR, which can further integrate federated learning and adversarial training for improved generalization. UniHAR is fully prototyped on the mobile platform and introduces low overhead to mobile devices. Extensive experiments demonstrate its superior performance in adapting HAR models across four open datasets.

Source: ACM Annual International Conference on Mobile Computing and Networking (MobiCom)

DOI: 10.1145/3570361.3613299

Authors: Huatao Xu, Pengfei Zhou, Rui Tan, Mo Li

Institutions: Nanyang Technological University, University of Pittsburgh, Hong Kong University of Science and Technology

Language: English

Topic: Context-Aware Activity Recognition Systems

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: bronze

Funders: National Research Foundation, National Research Foundation Singapore, Ministry of Education, India

Awards: G2072543325, G875168909