Hierarchical human activity recognition with fusion of audio and multiple inertial sensor modalities
Tuğce Alara Yılmaz, Hakan Yekta Yatbaz, Enver Ever, Adnan Yazici · 2025
Read Hierarchical human activity recognition with fusion of audio and multiple inertial sensor modalities by Tuğce Alara Yılmaz, Hakan Yekta Yatbaz, Enver Ever, Adnan Yazici (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12770422.txt.
Description
Tuğce Yılmaz, Hakan Yatbaz, Enver Ever, and Adnan Yazici sort daily activities from sound plus motion, in work from 2025. Activity sorting helps health care spot falls and odd breaks from normal routine. Phones and wearables can run such checks, but models must stay light on small devices.
Their model turns motion and audio streams into colour coded image inputs for learning. Layered sorting first splits broad groups, then splits fine acts inside each group. Tests used a hard public dataset with a layered activity structure. Balanced accuracy passed ninety percent, which beat prior work on the same data. Cpu and memory loads fell as well, which suits phone deployment.
Catalog record
| CREATOR | Tuğce Alara Yılmaz, Hakan Yekta Yatbaz, Enver Ever, Adnan Yazici |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 76 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |