通过几何内点判别实现激光雷达语义分割在损坏下的测试时训练
Test-Time Training for LiDAR Semantic Segmentation under Corruption via Geometric Inlier Discrimination
提出几何内点判别（GeoID）作为自监督辅助任务，通过注入合成离群点并区分几何一致的内点来驱动测试时训练，可复用于其他3D感知任务在分布偏移下的自适应。
原文依据 geometric inlier discrimination (GeoID), which injects synthetic off-manifold points into the input and trains the model to distinguish geometry-consistent inliers from synthetically displaced outliers
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语义分割必须在各种传感器和环境损坏下保持鲁棒性，才能在安全关键应用中可靠运行。现有的测试时自适应方法，包括基于伪标签和归一化统计的方法，已显示出有前景的结果，但在严重的分布偏移下仍可能表现不佳。为了补充这些方法，我们提出了一个几何感知的测试时训练框架，利用辅助自监督目标。我们的方法基于几何内点判别（GeoID），它向输入中注入合成的离流形点，并训练模型区分几何一致的内点与合成位移的离群点，从而能够在无标签测试数据上进行自适应。为了在真实损坏下进一步稳定这一过程，我们引入了双向不可靠点过滤（BiUPF），它使用源训练模型的内点分数来过滤原始点和合成点上的不可靠区域，将更新集中在高置信度样本上。在两个大规模损坏基准 SemanticKITTI-C 和 nuScenes-C 上的实验表明，我们的方法始终优于强测试时自适应基线，并在各种激光雷达损坏下提高了鲁棒性。我们的代码可在 https://github.com/hskim617/GeoID 获取。
英文原文摘要 1339 字符
LiDAR semantic segmentation must remain robust under various sensor and environmental corruptions to be reliable in safety-critical applications. Existing test-time adaptation methods, including approaches based on pseudo-labels and normalization statistics, have shown promising results but can still struggle under severe distribution shifts. To complement these approaches, we propose a geometry-aware test-time training framework that leverages an auxiliary self-supervised objective. Our method is based on geometric inlier discrimination (GeoID), which injects synthetic off-manifold points into the input and trains the model to distinguish geometry-consistent inliers from synthetically displaced outliers, enabling adaptation on unlabeled test data. To further stabilize this process under real corruptions, we introduce bidirectional unreliable point filtering (BiUPF), which uses inlier scores from the source-trained model to filter out unreliable regions on both original and synthetic points, focusing updates on high-confidence samples. Experiments on two large-scale corruption benchmarks, SemanticKITTI-C and nuScenes-C, show that our method consistently outperforms strong test-time adaptation baselines and improves robustness across diverse LiDAR corruptions. Our code is available at https://github.com/hskim617/GeoID.