Piezoelectric Acoustic Sensing for Sitting Pose Classification


Yuuki Shibuya (Tokyo University of Science), Go Irie (Tokyo University of Science)
The 35th British Machine Vision Conference
PDFPosterVideo (Right click to download)Supplementary

Abstract

Recognizing human poses, particularly under seated conditions, has become increasingly important due to the widespread adoption of remote work and teleconferencing. In this study, we present the first attempt to classify sitting poses using active acoustic sensing based on contact-type piezoelectric devices attached to a chair. Our framework analyzes multi-channel acoustic responses of known sweep signals transmitted through the chair and the seated body and captured by a piezoelectric microphone array. To enhance classification performance, we introduce two learning techniques tailored to this setting. Specifically, we introduce ChannelSwap (CS), a data augmentation method that leverages the geometric symmetry of the sensing system, and Symmetric Consistency Enhancement (SCE), a learning strategy designed to compensate for real-world symmetry deviations. Experiments on real sound data demonstrate that our method improves classification accuracy by 3.6\% compared to standard baselines, validating the feasibility and effectiveness of piezoelectric acoustic sensing for sitting pose classification. Resources are available at https://github.com/yuki10647/Piezoelectric-Acoustic-Sensing-for-Sitting-Pose-Classification.

Citation

@inproceedings{Shibuya_2025_BMVC,
author    = {Yuuki Shibuya and Go Irie},
title     = {Piezoelectric Acoustic Sensing for Sitting Pose Classification},
booktitle = {36th British Machine Vision Conference 2025, {BMVC} 2025, Sheffield, UK, November 24-27, 2025},
publisher = {BMVA},
year      = {2025},
url       = {https://bmva-archive.org.uk/bmvc/2025/assets/papers/Paper_55/paper.pdf}
}


Copyright © 2025 The British Machine Vision Association and Society for Pattern Recognition
The British Machine Vision Conference is organised by The British Machine Vision Association and Society for Pattern Recognition. The Association is a Company limited by guarantee, No.2543446, and a non-profit-making body, registered in England and Wales as Charity No.1002307 (Registered Office: Dept. of Computer Science, Durham University, South Road, Durham, DH1 3LE, UK).

Imprint | Data Protection