RPINet: Dual-branch attention-guided fusion of remote sensing images and point clouds for urban scene segmentation
Zhe Jing, Zhengguo Yan · 2026
Read RPINet: Dual-branch attention-guided fusion of remote sensing images and point clouds for urban scene segmentation by Zhe Jing, Zhengguo Yan (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-13196964.txt.
Description
Zhe Jing and Zhengguo Yan present a 2026 model called RPINet for labeling city scenes in three dimensions. Point clouds are sets of measured points that map streets and buildings. Labeling them is hard because points are sparse and uneven, and photos alone miss depth.
RPINet runs two branches side by side. One branch reads the three-dimensional points, while the other reads flat aerial images for texture and edges. An attention unit blends both streams, weighing each part by its value. On a large city test set, the model beat prior methods on overlap scores. It also held up on new data it had not seen. The approach may aid city planning and mapping.
Catalog record
| CREATOR | Zhe Jing, Zhengguo Yan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 68 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |