发表机构
University of Cyprus（塞浦路斯大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究在DALES、S3DIS、STPLS3D数据集上，针对KPConv和RandLA-Net架构，测试了6种重加权方案与5种感知不平衡损失，发现采样策略、不平衡程度及数据采集特征的相互作用影响激光雷达分割的不平衡缓解效果。
AI 中文摘要
激光雷达点云的类别不平衡问题给自主导航和城市制图中的语义分割带来了挑战。尽管二维视觉领域有大量不平衡缓解技术，但它们在三维场景中的有效性尚不明确。我们在三个数据集（DALES、S3DIS、STPLS3D）上，针对两种架构（KPConv、RandLA-Net），对六种重加权方案和五种感知不平衡的损失函数进行了基准测试。与均匀加权相比，逆频率加权的性能最多下降12%，且在少数类别中出现灾难性失效。对于结构化采样（KPConv），均匀加权的性能与复杂损失的差距在2%以内，但对于随机采样（RandLA-Net），其获益较少，差距最多达4.6%。损失景观分析揭示了复杂的相互作用：对于结构化采样，不平衡比例决定了真实激光雷达数据上的景观几何形状，但在合成数据上则与该几何形状解耦；对于随机采样，无论不平衡比例如何，景观对数据集几何形状均表现出高敏感性。针对所评估的两种基于点的架构，这些结果表明，采样策略（结构化与随机）、不平衡严重程度和数据采集特征之间的相互作用，决定了哪些缓解方法是有效的。
英文摘要
Class imbalance in LiDAR point clouds poses challenges for semantic segmentation in autonomous navigation and urban mapping. While 2D vision has numerous mitigation techniques, their effectiveness in 3D remains unclear. We benchmark six reweighting schemes and five imbalance-aware losses across three datasets (DALES, S3DIS, STPLS3D) using two architectures (KPConv, RandLA-Net). Inverse-frequency weighting degrades performance by up to 12% compared to uniform weighting, with catastrophic failures in minority classes. Uniform weighting performs within 2% of complex losses for structured sampling (KPConv) but benefits less for random sampling (RandLA-Net, up to 4.6% gap). Loss landscape analysis reveals a complex interplay: for structured sampling, imbalance ratio determines landscape geometry on real LiDAR data but decouples from it on synthetic data; for random sampling, landscapes show high sensitivity to dataset geometry regardless of imbalance ratio. For the two evaluated point-based architectures, these results suggest that the interaction between sampling strategy (structured vs. random), imbalance severity, and data acquisition characteristics shapes which mitigation approaches are effective.
Comments9 pages, 6 figures, IEEE International Conference on Image Processing (ICIP) 2026
DOI：10.1109/ICIP61757.2026.11630063
