WeatherMamba: Reliable Geometry-Aware Generalization for Adverse-Weather LiDAR Sensor Point-Cloud Semantic Segmentation
He Huang, Xintai Zhang, Yidan Zhang, Junxing Yang · 2026
Read WeatherMamba: Reliable Geometry-Aware Generalization for Adverse-Weather LiDAR Sensor Point-Cloud Semantic Segmentation by He Huang, Xintai Zhang, Yidan Zhang, Junxing Yang (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-13611661.txt.
Description
He Huang and three colleagues published a computing study in two thousand twenty-six on laser scanning in bad weather. Self-driving systems use LiDAR, which maps scenes with laser pulses. Rain, snow, and fog scatter those pulses and confuse scene labelling.
Their system, WeatherMamba, learns from clear-weather data yet aims to work in unseen storms. It judges how trusty each laser point looks, using signal strength and nearby point density. Points judged shaky get smaller corrections, without needing any weather report at run time. On standard test sets the system ranked near the top while using fewer resources than a rival design. Ablated versions scored worse, which supports the value of each added part.
Catalog record
| CREATOR | He Huang, Xintai Zhang, Yidan Zhang, Junxing Yang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |