Ensemble-based Semi-supervised Learning to Improve Noisy Soiling Annotations in Autonomous Driving

TL;DR

This paper presents an ensemble-based semi-supervised approach to improve noisy annotations of soiling in autonomous driving camera data, leading to better segmentation models and more accurate annotations.

Contribution

It introduces a pseudo-label driven ensemble method to identify and refine noisy annotations, enhancing model training on imperfect data.

Findings

Significant improvement in segmentation accuracy with refined labels

Effective identification and correction of problematic annotations

Demonstrated ability to refine coarse, low-cost annotations

Abstract

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.