Human Fall Detection with Infrared Imaging: A Comparison of Graph Convolutional Networks and YOLO
Karol Perliński, Artur Faltyński, Aleksandra Świetlicka · 2026
Read Human Fall Detection with Infrared Imaging: A Comparison of Graph Convolutional Networks and YOLO by Karol Perliński, Artur Faltyński, Aleksandra Świetlicka (2026) 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-13165531.txt.
Description
Perlinski, Faltynski, and Swietlicka set two AI ways to spot falls on heat shots. One way, GCN, reads bone maps as web forms for its picks. One way, YOLO, hunts forms in each shot frame on sight. Heat shots let the pair work in dark rooms with no lamp. The GCN form tops 99 percent on its sort scores. It reads both flat and deep bone maps in the runs.
YOLOv8 gets timed for live spot work on heat frames. Each way has clear strong points in the head-to-head runs. YOLO pins where a form sits in the frame fast. GCN reads how the form moves through the fall path. A mixed frame pairs both skills for field use next. Old-age care and ward watch stand as the use case.
Catalog record
| CREATOR | Karol Perliński, Artur Faltyński, Aleksandra Świetlicka |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |