.png){kind=icon}
.png){kind=icon}
Return
CoSS: Co-optimizing sensor and sampling rate for data-efficient human activity recognition
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2025.129959.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Proposes CoSS, a novel framework to co-optimize sensor modality and sampling rates. • Introduces trainable weight scores for efficient sensor and sampling rate selection. • Demonstrates significant hardware and computational efficiency in HAR systems. • Validates effectiveness using multiple public HAR benchmark datasets.
Journal
IF:
7.5
Papers:
2.9W
Citations:
10.2W