Manual annotation of soiling on surround view cameras is a very challenging and expensive task. The unclear boundary for various soiling categories like water drops or mud particles usually results in a large variance in the annotation quality. As a result, the models trained on such poorly annotated data are far from being optimal. In this paper, we focus on handling such noisy annotations via pseudo-label driven ensemble model which allow us to quickly spot problematic annotations and in most cases also sufficiently fixing them. We train a soiling segmentation model on both noisy and refined labels and demonstrate significant improvements using the refined annotations. It also illustrates that it is possible to effectively refine lower cost coarse annotations.
Ensemble-Based Semi-Supervised Learning to Improve Noisy Soiling Annotations in Autonomous Driving
2021-09-19
1783406 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Semi-supervised deep ensemble learning for travel mode identification
Elsevier | 2020
|British Library Online Contents | 2016
|Using Predictive Uncertainty for Cleaning Noisy Annotations
Deutsches Zentrum für Luft- und Raumfahrt (DLR) | 2022
|