Dynamic multi-scale fusion network for road damage detection in complex street views
Hong Zhou, Yiyang Qin, Jiahuan Ren, Yan Zhang, Shangbing Gao · 2026
Read Dynamic multi-scale fusion network for road damage detection in complex street views by Hong Zhou, Yiyang Qin, Jiahuan Ren, Yan Zhang, Shangbing Gao (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/PMC13507222.txt.
Description
Hong Zhou, Yiyang Qin, Jiahuan Ren, Yan Zhang, and Shangbing Gao present 2026 work on a vision network that spots road damage in messy street photos. Damage ranges from hairline cracks to wide potholes, and thin cracks curve in irregular ways, which defeats single-scale detectors. Their network builds on a fast YOLO detector with new fusion blocks in its backbone and neck.
One block mixes grouped convolutions across scales to catch damage of every size. A second branch set uses snake-shaped convolutions that follow curving cracks, plus a light attention step for merging. Tests on three public road datasets showed a strong balance of accuracy and speed. Cities could survey pavement more cheaply from street imagery.
Catalog record
| CREATOR | Hong Zhou, Yiyang Qin, Jiahuan Ren, Yan Zhang, Shangbing Gao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56953 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |