Failure or Drift? Evaluating Monocular SLAM under Synthetic and Real-World Corruptions
2026-08-31 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionRobotics
AI summaryⓘ
The authors studied how well monocular Visual SLAM systems work when the input images are corrupted by things like fog, rain, or bad lighting. They tested both traditional feature-based methods and newer learned trackers under controlled but realistic simulated conditions and compared this to real-world adverse weather data. They found that learned methods tend to avoid complete failure but instead gradually drift away from the correct position. Also, more physically realistic corruptions better predicted the systems' real-world performance than simple image changes like brightness. This suggests that evaluating SLAM robustness needs careful choice of test conditions.
Visual SLAMmonocular SLAMfeature-based SLAMlearned trackersadverse weatherimage corruptiontrajectory driftsynthetic stress testsrobustness evaluation4Seasons dataset
Authors
Abhay Skaria Thomas, Shashank Agnihotri, Margret Keuper
Abstract
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.