Human motion recognition based on feature fusion and residual networks
Xiaoyu Luo, Qiusheng Li · 2024
Read Human motion recognition based on feature fusion and residual networks by Xiaoyu Luo, Qiusheng Li (2024) 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-11586430.txt.
Description
Xiaoyu Luo and Qiusheng Li raise the accuracy of human motion sensing by fusing two radar views. Radar captures of a moving body carry rich clues, but a single feature often reads them too weakly. Their method pairs frequency modulated continuous wave radar with a residual network.
In the 2024 paper, echo signals from six kinds of motion are turned into range-time and Doppler-time maps. A dual-channel network then reads both maps at once. The design upgrades the residual blocks with inception modules and adds an attention block for sharper feature picking. Tests show the fused model beats single-feature versions by a clear margin, proof that the two views correct each other's blind spots.
Catalog record
| CREATOR | Xiaoyu Luo, Qiusheng Li |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |