To Örebro University

oru.seÖrebro University Publications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Siamese masked reconstruction with temporal rotation consistency for inertial-based human activity recognition
Örebro University, School of Science and Technology.ORCID iD: 0000-0001-6647-4215
Örebro University, School of Science and Technology.ORCID iD: 0000-0002-6013-4874
Human Robotics, DFESTS, University of Alicante, Alicante, Spain; MAD-Robotics, Escuela Técnica Superior de Ingeniería y Diseño Industrial (ETSIDI), Universidad Politécnica de Madrid, Madrid, Spain.ORCID iD: 0000-0002-3908-4921
2026 (English)In: Scientific Reports, E-ISSN 2045-2322, Vol. 16, no 1, article id 25518Article in journal (Refereed) Published
Abstract [en]

Inertial-based Human Activity Recognition (HAR) aims at inferring activities performed by an individual from data on wearable inertial sensors. However, HAR often suffers from poor generalization to unseen users due to data heterogeneity across different individuals and conditions. While collecting and curating large labeled datasets could help alleviate this limitation, it is also costly. To reduce this gap without relying on extensive annotated data, we propose a self-supervised framework grounded in how IMU signals transform under rotation: applying a rotation matrix changes how motion projects onto the sensor axes without altering the underlying motion. A model trained to align representations of a signal and its rotated counterpart is therefore encouraged to capture motion content rather than sensor-frame patterns. We exploit this property by training a Siamese masked convolutional autoencoder that learns by reconstructing masked inputs while aligning representations of original signals and its rotated version, further regularized by a temporal consistency loss that enforces agreement between the temporal structure of original and rotated representations within each window. We evaluate on four public HAR benchmarks covering diverse scenarios, including activities of the daily living (ADL) and sports activities, using cross-subject evaluation, where our method yields improvements of +1.4 and +1.2 percentage points over the strongest baseline with linear and MLP probes respectively, averaged across datasets and sensor positions. We further show that our method obtains competitive results against the fully supervised baseline in low-data regimes. The code is available on https://github.com/FranciscoCalatrava/Rotation_Siamese_Masked.git .

Place, publisher, year, edition, pages
Nature Portfolio, 2026. Vol. 16, no 1, article id 25518
National Category
Computer graphics and computer vision
Identifiers
URN: urn:nbn:se:oru:diva-130676DOI: 10.1038/s41598-026-64846-5ISI: 001850341100008PubMedID: 42601386OAI: oai:DiVA.org:oru-130676DiVA, id: diva2:2092590
Funder
Örebro UniversityWallenberg AI, Autonomous Systems and Software Program (WASP)Knut and Alice Wallenberg FoundationAvailable from: 2026-08-17 Created: 2026-08-17 Last updated: 2026-08-31Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textPubMed

Authority records

Calatrava Nicolás, Francisco M.Stoyanov, TodorMartinez Mozos, Oscar

Search in DiVA

By author/editor
Calatrava Nicolás, Francisco M.Stoyanov, TodorMartinez Mozos, Oscar
By organisation
School of Science and Technology
In the same journal
Scientific Reports
Computer graphics and computer vision

Search outside of DiVA

GoogleGoogle Scholar

doi
pubmed
urn-nbn

Altmetric score

doi
pubmed
urn-nbn
Total: 15 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf