发表机构
University of Mannheim; Max-Planck-Institute for Informatics（曼海姆大学; 马克斯·普朗克信息学研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究评估单目SLAM在合成与真实世界干扰下的表现，区分跟踪失效与漂移，发现学习型跟踪器会产生持续漂移，且其排序随干扰物理保真度变化。
AI 中文摘要
视觉SLAM通常在干净轨迹上进行评估，尽管部署失效常由恶劣天气、光照、模糊和传感器伪影导致。受控干扰因能隔离这些因素颇具吸引力，但合成压力测试仅当它得出与所近似条件相同的工程结论时才有用。本工作针对单目SLAM研究该问题，评估经典基于特征的系统和两个学习型跟踪器在图像空间、几何感知及复合干扰下的表现，并将其与4Seasons真实世界干扰下的行为对比。评估未仅将鲁棒性降低归结为单一轨迹误差，而是区分了显式跟踪失效与保持运行的方法累积的漂移。结果显示，学习型跟踪器大多将灾难性损失替换为持续且有时严重的漂移；更重要的是，学习系统的表观排序随干扰的物理保真度变化：结构化雨和雾代理保留真实世界排序，而简单光照代理则不。代码可在：this https URL获取。
英文摘要
Visual SLAM is commonly evaluated on clean trajectories, although deployment failures are often caused by adverse weather, illumination, blur, and sensor artifacts. Controlled corruptions are attractive because they isolate such factors, but a synthetic stress test is useful only when it leads to the same engineering conclusion as the condition it is intended to approximate. This work examines that question for monocular SLAM. We evaluate a classical feature-based system and two learned trackers under image-space, geometry-aware, and compound corruptions, and compare their behavior with adverse conditions from 4Seasons. Rather than reducing robustness to a single trajectory error, the evaluation separates explicit tracking failure from drift accumulated by methods that remain active. The results show that learned trackers largely replace catastrophic loss with sustained, and sometimes severe, drift. More importantly, the apparent ordering of the learned systems changes with the physical fidelity of the corruption: structured rain and fog proxies preserve the real-world ordering, whereas a simple illumination proxy does not. Code is available at: https://github.com/abhaythomas/master_thesis_vslamlab_robustness.
CommentsAccepted at the 3rd NeuSLAM workshop at ECCV 2026
