Metraipolis › archive › dc_45d7483080da

Coverage-Aware Virtual IMU Augmentation for Low-Resource Human Activity Recognition — arXiv:2609.16768

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.16768
License (as archived): CC0-1.0
Archived: 2026-10-15T18:32:12.669Z
SHA-256 of the archived text: 7a9f4fa3dafba4b5ac39922adc5ba1badf7d7eb304b04884bad9c4212d2f11aa
Stable URL of this page: https://metraipolis.com/documento/dc_45d7483080da

How to cite

Coverage-Aware Virtual IMU Augmentation for Low-Resource Human Activity Recognition — arXiv:2609.16768 — arXiv (en), https://arxiv.org/abs/2609.16768 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_45d7483080da (Metraipolis archive dc_45d7483080da, sha256 7a9f4fa3dafba4b5ac39922adc5ba1badf7d7eb304b04884bad9c4212d2f11aa)

BibTeX

@misc{dc45d7483080da,
  title = {Coverage-Aware Virtual IMU Augmentation for Low-Resource Human Activity Recognition — arXiv:2609.16768},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.16768},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_45d7483080da; sha256 7a9f4fa3dafba4b5ac39922adc5ba1badf7d7eb304b04884bad9c4212d2f11aa},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.16768. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

IMU-based human activity recognition (HAR) enables continuous, privacy-friendly monitoring of daily activities using wearable sensors. However, building reliable HAR models that generalize across diverse users and real-world conditions requires large amounts of labeled IMU data, which are expensive and difficult to collect. Existing approaches mainly rely on augmentation or synthesis to expand available data, but indiscriminately adding virtual samples may provide little new coverage and introduce unreliable supervision. To overcome these challenges, we propose a novel coverage-aware virtual IMU augmentation framework that decides where to supplement real data, how to generate and select virtual candidates, and how strongly to weight them during training. Specifically, we select diversity and scarcity anchors in a learned sensor embedding space, convert anchor dynamics into prompts, and generate virtual IMU candidates for each anchor. We then rank candidates by a selection cost combining anchor proximity and label consistency, and incorporate the selected candidates into HAR training with reliability-based weights. Experiments on public HAR benchmarks show that our method consistently improves recognition performance over competitive baselines, and ablation studies confirm the effectiveness of the proposed framework design.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-16 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.16768 — 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 2026-10-15T18:03:09.470Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-16?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).