LHA-YOLO: A Lightweight and High-Accuracy Detector via Parallel Attention and Divide-and-Conquer Fusion for UAV Images
Jianxiu Yang, Xiong Pan, Qingzhe Pan · 2026
Read LHA-YOLO: A Lightweight and High-Accuracy Detector via Parallel Attention and Divide-and-Conquer Fusion for UAV Images by Jianxiu Yang, Xiong Pan, Qingzhe Pan (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-13210410.txt.
Description
Jianxiu Yang, Xiong Pan, and Qingzhe Pan sharpen drone vision for tiny objects in a 2026 paper. From a drone's height, cars and people shrink to a few pixels against cluttered ground. Standard detectors blur meaning and position cues together when merging image scales.
Their network, LHA-YOLO, splits the work. One light module attends to space and channel cues in parallel. A divide-and-conquer path then handles meaning flowing down and fine detail flowing up through separate guided modules. On two drone benchmarks it beat its base model on accuracy scores while cutting computing load, a useful mix for live flight hardware.
Catalog record
| CREATOR | Jianxiu Yang, Xiong Pan, Qingzhe Pan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |