FloorYOLO: an attention-enhanced YOLOv8n for floor plan object recognition
S. Santhosh, M. S. Rashmika, A. Sasithradevi, Sabari Nathan · 2026
Read FloorYOLO: an attention-enhanced YOLOv8n for floor plan object recognition by S. Santhosh, M. S. Rashmika, A. Sasithradevi, Sabari Nathan (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-13619391.txt.
Description
Santhosh, Rashmika, Sasithradevi and Sabari Nathan present FloorYOLO, a detector for objects in floor plan images. Plans pack doors, walls, windows and room types close together, with like symbols that vary and blur at the edges. Older rule-based and learning methods struggle with those traits.
FloorYOLO builds on YOLOv8n and adds an Efficient Channel Attention module that sharpens features at low extra cost. A PAN neck joins fine spatial detail with semantic cues before the head makes detections. On the sixteen-class SESYD set it reached a mean average precision of 0.9794 on the strict range, and on real-world CVC-FP it beat the baseline score.
Catalog record
| CREATOR | S. Santhosh, M. S. Rashmika, A. Sasithradevi, Sabari Nathan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 73 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |