Lightweight personalisation for MEMS-based wearables : a padel stroke recognition case study
(2026)- Abstract
- This work investigates lightweight personalisation of micro-electro-mechanical systems (MEMS)-based wearables using a public padel database as a case study. We compare a centralised CNN model, single-user models and two fine-tuning schemes (full and last-layer) on wrist-worn IMU data from 23 players and 13 stroke classes. Personalised models with data augmentation achieve weighted F1-scores above 90%, closing most of the gap to an optimistic single-user upper bound while reducing inter-subject variability. FLOP and memory analyses show that last-layer fine-tuning offers a favorable trade-off between accuracy and efficiency for on-device deployment in MEMS-based wearables.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/a64e6e20-0494-4c5e-999e-c93b3cdd6aca
- author
- Gascon, Alberto
; Akbarian, Fatemeh
LU
; Aminifar, Amir
LU
; Marco, Alvaro
and Casas, Roberto
- organization
-
- LTH Profile Area: AI and Digitalization
- Secure and Networked Systems
- ELLIIT: the Linköping-Lund initiative on IT and mobile communication
- LTH Profile Area: Engineering Health
- NEXTG2COM – a Vinnova Competence Centre in Advanced Digitalisation
- LTH Profile Area: Water
- LU Profile Area: Natural and Artificial Cognition
- publishing date
- 2026
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- in press
- subject
- host publication
- European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
- pages
- 6 pages
- language
- English
- LU publication?
- yes
- id
- a64e6e20-0494-4c5e-999e-c93b3cdd6aca
- date added to LUP
- 2026-03-25 16:35:37
- date last changed
- 2026-03-26 10:04:36
@inproceedings{a64e6e20-0494-4c5e-999e-c93b3cdd6aca,
abstract = {{This work investigates lightweight personalisation of micro-electro-mechanical systems (MEMS)-based wearables using a public padel database as a case study. We compare a centralised CNN model, single-user models and two fine-tuning schemes (full and last-layer) on wrist-worn IMU data from 23 players and 13 stroke classes. Personalised models with data augmentation achieve weighted F1-scores above 90%, closing most of the gap to an optimistic single-user upper bound while reducing inter-subject variability. FLOP and memory analyses show that last-layer fine-tuning offers a favorable trade-off between accuracy and efficiency for on-device deployment in MEMS-based wearables.}},
author = {{Gascon, Alberto and Akbarian, Fatemeh and Aminifar, Amir and Marco, Alvaro and Casas, Roberto}},
booktitle = {{European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning}},
language = {{eng}},
title = {{Lightweight personalisation for MEMS-based wearables : a padel stroke recognition case study}},
year = {{2026}},
}