Unsupervised Diffusion Model for Sensor-based Human Activity Recognition

Si Zuo, Vitor Fortes, Sungho Suh, Stephan Sigg, Paul Lukowicz

Research output: Contribution to conferenceAbstractScientificpeer-review

Abstract

Recognizing human activities from sensor data is a vital task in various domains, but obtaining diverse and labeled sensor data remains challenging and costly. In this paper, we propose an unsupervised statistical feature-guided diffusion model for sensor-based human activity recognition. The proposed method aims to generate synthetic time-series sensor data without relying on labeled data, addressing the scarcity and annotation difficulties associated with real-world sensor data. By conditioning the diffusion model on statistical information such as mean, standard deviation, Z-score, and skewness, we generate diverse and representative synthetic sensor data. We conducted experiments on public human activity recognition datasets and compared the proposed method to conventional oversampling methods and state-of-the-art generative adversarial network methods. The experimental results demonstrate that the proposed method can improve the performance of human activity recognition and outperform existing techniques.
Original languageEnglish
Pages205
Number of pages1
DOIs
Publication statusPublished - 2023
MoE publication typeNot Eligible
EventACM International Joint Conference on Pervasive and Ubiquitous Computing and ACM International Symposium on Wearable Computers - Cancun, Mexico
Duration: 8 Oct 202312 Oct 2023

Conference

ConferenceACM International Joint Conference on Pervasive and Ubiquitous Computing and ACM International Symposium on Wearable Computers
Abbreviated titleUbiComp/ISWC
Country/TerritoryMexico
CityCancun
Period08/10/202312/10/2023

Fingerprint

Dive into the research topics of 'Unsupervised Diffusion Model for Sensor-based Human Activity Recognition'. Together they form a unique fingerprint.

Cite this