End to end polysemantic cooperative mixed task trainer for UAV target detection
Xueying Liao, Xingran Guo, Askar Rozi, Haizheng Yu, Abdukerim Haji · 2024
Read End to end polysemantic cooperative mixed task trainer for UAV target detection by Xueying Liao, Xingran Guo, Askar Rozi, Haizheng Yu, Abdukerim Haji (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-11608260.txt.
Description
Drone cameras spot small urban targets better with a meaning-aware transformer, Xueying Liao and co-authors present in a 2024 paper. City scenes from above mix subtle targets with cluttered backgrounds. Their detector guides attention with context drawn from the whole scene.
A mixed-task training scheme adds helper heads with varied labeling rules, which stretches the encoder's learning. Queries are tailored to the scheme's assignments. Slanted drone shots also feed precise 3D rebuilds that aid detection. The goal is sharper finds of vehicles, people, and structures in crowded aerial views, where small size and shadow conspire against standard detectors.
Catalog record
| CREATOR | Xueying Liao, Xingran Guo, Askar Rozi, Haizheng Yu, Abdukerim Haji |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 219 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |