DDRN: DETR with dual refinement networks for autonomous vehicle object detection
Jiayao Li, Chak Fong Cheang, Zhaolong Du, Xiaoyuan Yu, Suigu Tang, Qianxiang Cheng · 2025
Read DDRN: DETR with dual refinement networks for autonomous vehicle object detection by Jiayao Li, Chak Fong Cheang, Zhaolong Du, Xiaoyuan Yu, Suigu Tang, Qianxiang Cheng (2025) 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-12738654.txt.
Description
Self-driving cars must spot small and distant objects reliably, and cluttered street scenes make that hard. Jiayao Li and co-authors present DDRN in this 2025 paper, a detection system that adds two refinement stages on top of a modern transformer-based detector.
One upstream tweak reweights how deep features teach shallow ones, cutting background mistakes on small objects. Two later networks then polish the class guesses and the box positions separately. On two public driving data sets the full system beat strong rivals by about a point of average precision. Wide tests back its steadiness across varied street scenes.
Catalog record
| CREATOR | Jiayao Li, Chak Fong Cheang, Zhaolong Du, Xiaoyuan Yu, Suigu Tang, Qianxiang Cheng |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |