TerraSeg：面向任意激光雷达的自监督地面分割
TerraSeg: Self-Supervised Ground Segmentation for Any LiDAR
提出一种无需人工标注、跨传感器配置的激光雷达地面分割自监督框架，其 PseudoLabeler 通过逐扫描运行时优化生成高质量伪标签，可复用于其他点云分割任务的标签生成与域泛化。
原文依据 we propose PseudoLabeler, a novel module that generates high-quality ground and non-ground labels through self-supervised per-scan runtime optimization.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论激光雷达感知是机器人技术的基础，使机器能够在三维环境中理解周围环境。基于激光雷达的场景理解与导航的一项关键任务是地面分割。然而，现有方法要么针对特定传感器配置手工设计，要么依赖昂贵的逐点人工标注，严重限制了其泛化性与可扩展性。为克服这一局限，我们提出 TerraSeg，这是首个用于激光雷达地面分割的自监督、域无关模型。我们在 OmniLiDAR 上训练 TerraSeg，这是一个统一的大规模数据集，聚合并标准化了来自 12 个主要公开基准的数据。OmniLiDAR 涵盖 15 种不同传感器模型的近 2200 万次原始扫描，为学习高度可泛化的地面模型提供了前所未有的多样性。为了在无人工标注的情况下监督训练，我们提出 PseudoLabeler，这是一种新颖的模块，通过自监督的逐扫描运行时优化生成高质量的地面与非地面标签。大量评估表明，尽管未使用任何人工标签，TerraSeg 在 nuScenes、SemanticKITTI 和 Waymo Perception 上达到了最优结果，同时实现了实时性能。我们的代码与模型权重已公开。
英文原文摘要 1249 字符
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.