FugSeg: Fast Uncertainty-aware Ground Segmentation for 3D Point Cloud

TL;DR

FugSeg is a fast, uncertainty-aware ground segmentation method for 3D LiDAR point clouds that improves accuracy and speed across various environments, suitable for resource-limited systems.

Contribution

It introduces a generalizable polar grid map, an adaptive slope with uncertainty modeling, and a fine-grained elevation estimation for robust ground segmentation.

Findings

Outperforms state-of-the-art non-learning methods in accuracy and F1 score.

Achieves real-time processing at 135 Hz and 487 Hz on CPU.

Effectively handles reflection noise and isolated ground in diverse environments.

Abstract

In LiDAR-based environment perception systems, ground segmentation is a key preprocessing step supporting various applications such as mapping and navigation. Although extensively studied, problems such as reflection noise and isolated ground remain challenging. To address these issues, we propose FugSeg, a fast uncertainty-aware ground segmentation method. A polar grid map is adopted as the point cloud representation to ensure generalizability across LiDAR types. Building on that, we develop a within- and cross-segment ground labeling strategy that identifies not only directly visible ground cells but also those that are isolated or occluded. During this process, an adaptive slope is introduced, which incorporates measurement uncertainties to enhance its reliability under complex terrain. Finally, to achieve point-level ground segmentation, a fine-grained ground elevation estimation method is introduced. Throughout the complete workflow, reflection noise is explicitly handled via the proposed noisy ground cells. We conduct comprehensive evaluations on four public datasets covering both structured and unstructured environments. Results show that FugSeg outperforms state-of-the-art non-learning methods, achieving the highest F1, accuracy, and mIoU across all datasets, while maintaining the fastest runtime (135 Hz and 487 Hz for 64- and 32-layer LiDARs) using a single CPU thread, making it suitable for resource-limited systems. The code will be available at https://github.com/Leo-YuLi/FugSeg.