Bridging the Scale Gap: A Multi-Scale Feature Enhancement Framework for UAV Aerial Image Object Detection
Dan Shan, Zanqi Qiu, Dadi Cai, Carlos Ferran, Xuan Tong, Dongming Liu · 2026
Read Bridging the Scale Gap: A Multi-Scale Feature Enhancement Framework for UAV Aerial Image Object Detection by Dan Shan, Zanqi Qiu, Dadi Cai, Carlos Ferran, Xuan Tong, Dongming Liu (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-13611273.txt.
Description
Dan Shan and five co-authors present a 2026 framework for finding objects in drone images. Aerial photos mix tiny and large objects, dense clusters, and busy backgrounds, which strain standard detectors. Transformer detectors model the whole scene well but can lose fine detail and cross scale links.
Their system, MSF-DETR, builds on an existing detector and adds four linked parts. These parts improve cross scale contact, sharpen channel choice, keep fine spatial detail, and steady the training signal in dense scenes. On two public benchmarks it beat its base model by several points on standard accuracy scores. The authors argue that coordinating the parts, rather than stacking separate modules, brings the gain in robustness.
Catalog record
| CREATOR | Dan Shan, Zanqi Qiu, Dadi Cai, Carlos Ferran, Xuan Tong, Dongming Liu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 98 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |