TerraSeg: Self-Supervised Ground Segmentation for Any LiDAR
This work solves the generalization and scalability problem in LiDAR ground segmentation for robotics by eliminating the need for sensor-specific tuning or manual annotations.
TerraSeg introduces the first self-supervised, domain-agnostic LiDAR ground segmentation model, trained on a diverse dataset of 22 million scans from 15 sensor models. It achieves state-of-the-art results on nuScenes, SemanticKITTI, and Waymo Perception without manual labels, with real-time performance.
LiDAR perception is fundamental to robotics, enabling machines to understand their environment in 3D. A crucial task for LiDAR-based scene understanding and navigation is ground segmentation. However, existing methods are either handcrafted for specific sensor configurations or rely on costly per-point manual labels, severely limiting their generalization and scalability. To overcome this, we introduce TerraSeg, the first self-supervised, domain-agnostic model for LiDAR ground segmentation. We train TerraSeg on OmniLiDAR, a unified large-scale dataset that aggregates and standardizes data from 12 major public benchmarks. Spanning almost 22 million raw scans across 15 distinct sensor models, OmniLiDAR provides unprecedented diversity for learning a highly generalizable ground model. To supervise training without human annotations, we propose PseudoLabeler, a novel module that generates high-quality ground and non-ground labels through self-supervised per-scan runtime optimization. Extensive evaluations demonstrate that, despite using no manual labels, TerraSeg achieves state-of-the-art results on nuScenes, SemanticKITTI, and Waymo Perception while delivering real-time performance. Our code and model weights are publicly available.