Towards Generalized Range-View LiDAR Segmentation in Adverse Weather

TL;DR

This paper introduces a modular framework that improves the robustness of range-view LiDAR segmentation under adverse weather by processing geometric and reflectance features separately, enhancing real-world applicability.

Contribution

It proposes a lightweight, modular approach with novel modules for geometric abnormality suppression and reflectance calibration to enhance weather robustness without changing core models.

Findings

Significant improvement in adverse weather conditions

Minimal impact on inference speed

Effective across multiple benchmarks and models

Abstract

LiDAR segmentation has emerged as an important task to enrich scene perception and understanding. Range-view-based methods have gained popularity due to their high computational efficiency and compatibility with real-time deployment. However, their generalized performance under adverse weather conditions remains underexplored, limiting their reliability in real-world environments. In this work, we identify and analyze the unique challenges that affect the generalization of range-view LiDAR segmentation in severe weather. To address these challenges, we propose a modular and lightweight framework that enhances robustness without altering the core architecture of existing models. Our method reformulates the initial stem block of standard range-view networks into two branches to process geometric attributes and reflectance intensity separately. Specifically, a Geometric Abnormality Suppression (GAS) module reduces the influence of weather-induced spatial noise, and a Reflectance Distortion Calibration (RDC) module corrects reflectance distortions through memory-guided adaptive instance normalization. The processed features are then fused and passed to the original segmentation pipeline. Extensive experiments on different benchmarks and baseline models demonstrate that our approach significantly improves generalization to adverse weather with minimal inference overhead, offering a practical and effective solution for real-world LiDAR segmentation.