Albazzai, Norah, Rana, Omer  ORCID: https://orcid.org/0000-0003-3597-2646 and Perera, Charith ORCID: https://orcid.org/0000-0002-0190-3346
2026.
SenseLess: Minimal vision, maximum insight for smart homes.
Presented at: 2026 IEEE International Conference on Pervasive Computing and Communications (PerCom),
Pisa, Italy,
16-20 March 2026.
2026 IEEE International Conference on Pervasive Computing and Communications (PerCom).
IEEE,
pp. 1-10.
10.1109/PerCom67906.2026.11524273
|
Preview |
PDF
- Accepted Post-Print Version
Available under License Creative Commons Attribution. Download (1MB) | Preview |
Abstract
We present SenseLess, a hybrid anomaly detection framework for smart homes that, during the training phase, automatically labels images without manual annotation by combining sensor-guided detection, self-supervised visual clustering, and unsupervised multi-sensor delay estimation for precise alignment. During operation, the system relies primarily on non-vision sensors and activates a confidence-aware vision model only under low-confidence, thereby preserving privacy while maintaining adaptability. Evaluated in real home monitoring, SenseLess achieved an average label coverage of 97.65% with 94.9% accuracy and reduced vision usage to less than 4% of wall-clock operating time. Calibration mechanisms and minimal configuration requirements support scalability and deployment across diverse residential environments.
| Item Type: | Conference or Workshop Item - published (Paper) |
|---|---|
| Date Type: | Publication |
| Status: | Published |
| Schools: | Schools > Computer Science & Informatics |
| Additional Information: | RRS policy applied |
| Publisher: | IEEE |
| ISBN: | 9798331576134 |
| ISSN: | 2474-249X |
| Date of First Compliant Deposit: | 3 June 2026 |
| Last Modified: | 05 Aug 2026 00:00 |
| URI: | https://orca.cardiff.ac.uk/id/eprint/187312 |
Actions (repository staff only)
 |
Edit Item |
Altmetric
Altmetric