TinyDark-YOLO for adaptive and lightweight object detection in low-light conditions
Xuefang Zheng, Lifang Chen, Jiechi Zhang · 2026
Read TinyDark-YOLO for adaptive and lightweight object detection in low-light conditions by Xuefang Zheng, Lifang Chen, Jiechi Zhang (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/PMC13482588.txt.
Description
Xuefang Zheng and co-authors present a 2026 lean detector for spotting objects in near-dark photos, built on a recent YOLO model. Dim scenes crush contrast and bury small targets in noise, while brightening photos first eats computing power. Their design adds a self-tuning brightness step, a context module that links distant parts of the frame, and a slim detection head that hunts weak traces.
On a standard dark-photo dataset the model scored higher than its parent design by over two points on the main accuracy scale. Computing load dropped at the same time, from the parent level to a leaner figure. A second face dataset in the dark gave competitive results as well. Night driving, watch cameras, and drone checks all need this balance of sharp eyes and light maths.
Catalog record
| CREATOR | Xuefang Zheng, Lifang Chen, Jiechi Zhang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 81986 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |