Soiling detection for Advanced Driver Assistance Systems
This work addresses robustness in advanced driver assistance systems for automotive applications, though it appears incremental as it focuses on dataset refinement and method comparison rather than introducing a fundamentally new approach.
The paper tackles soiling detection for automotive cameras by framing it as a semantic segmentation problem, comparing popular segmentation methods against tile-level classification approaches and showing their superiority in performance. It also identifies data-leakage and imprecise annotations in the Woodscape dataset, creating a smaller subset that enables comparable results in much shorter time.
Soiling detection for automotive cameras is a crucial part of advanced driver assistance systems to make them more robust to external conditions like weather, dust, etc. In this paper, we regard the soiling detection as a semantic segmentation problem. We provide a comprehensive comparison of popular segmentation methods and show their superiority in performance while comparing them to tile-level classification approaches. Moreover, we present an extensive analysis of the Woodscape dataset showing that the original dataset contains a data-leakage and imprecise annotations. To address these problems, we create a new data subset, which, despite being much smaller, provides enough information for the segmentation method to reach comparable results in a much shorter time. All our codes and dataset splits are available at https://github.com/filipberanek/woodscape_revision.