Metraipolis › archive › dc_e37b2bd5ca79

TransfHAR: Self-Supervised Wrist Representations for On-Demand Activity Recognition — arXiv:2608.15861

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.15861
License (as archived)
CC0-1.0
Archived
2026-10-18T19:41:05.164Z
SHA-256 of the archived text
4c5e9219b9a52462897f79b0b291f0d507318ddee68727b8a7afa5e9044e9d12
Stable URL of this page
https://metraipolis.com/documento/dc_e37b2bd5ca79

How to cite

TransfHAR: Self-Supervised Wrist Representations for On-Demand Activity Recognition — arXiv:2608.15861 — arXiv (en), https://arxiv.org/abs/2608.15861 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e37b2bd5ca79 (Metraipolis archive dc_e37b2bd5ca79, sha256 4c5e9219b9a52462897f79b0b291f0d507318ddee68727b8a7afa5e9044e9d12)
BibTeX
@misc{dce37b2bd5ca79,
  title = {TransfHAR: Self-Supervised Wrist Representations for On-Demand Activity Recognition — arXiv:2608.15861},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.15861},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e37b2bd5ca79; sha256 4c5e9219b9a52462897f79b0b291f0d507318ddee68727b8a7afa5e9044e9d12},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.15861. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Fine-grained wrist activity recognition can support applications such as procedural step guidance and context-aware assistance, yet acquiring labeled data for every new task, user, and activity granularity remains a bottleneck. We present TransfHAR, a self-supervised wrist IMU framework for on-demand, fine-grained activity recognition by learning transferable motion priors from global, unlabeled activities. We show that self-supervised pretraining on coarse wrist IMU activities (e.g., sitting, walking, exercise) learns motion structure rich enough to transfer to fine-grained manipulative, gestural, and procedural activities (e.g., snapping, stirring, waving) that are absent from pretraining. We implement TransfHAR as a real-time smartwatch application that lets users define and expand their own activity set for personalized recognition from only a few demonstrations. Across three offline cross-dataset evaluations, TransfHAR matches or exceeds fully supervised baselines that use complete label sets with equal or additional sensor channels, by 6.2 balanced-accuracy points on average. In an in-lab study with 10 participants each performing seven novel wrist activities, TransfHAR reaches 86.7% balanced accuracy across participants with five examples per class and 90.4% when updated from a single one-minute recording per class. These results indicate that broad self-supervised wrist pretraining provides an effective foundation for on-demand fine-grained activity recognition.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.15861 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-18T19:02:44.130Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).