TiledSoilingNet: Tile-level Soiling Detection on Automotive Surround-view Cameras Using Coverage Metric

TL;DR

This paper introduces TiledSoilingNet, a fast and efficient tile-level soiling detection method for automotive surround-view cameras, enabling automatic cleaning triggers and partial functionality in autonomous vehicles.

Contribution

It presents a novel coverage regression approach for soiling detection that is faster and more suitable for low-power embedded systems than traditional segmentation methods.

Findings

Coverage regression outperforms dominant class learning in mixed soiling scenarios.

The proposed decoder is an order of magnitude faster than segmentation decoders.

Integrated multi-task model improves detection efficiency.

Abstract

Automotive cameras, particularly surround-view cameras, tend to get soiled by mud, water, snow, etc. For higher levels of autonomous driving, it is necessary to have a soiling detection algorithm which will trigger an automatic cleaning system. Localized detection of soiling in an image is necessary to control the cleaning system. It is also necessary to enable partial functionality in unsoiled areas while reducing confidence in soiled areas. Although this can be solved using a semantic segmentation task, we explore a more efficient solution targeting deployment in low power embedded system. We propose a novel method to regress the area of each soiling type within a tile directly. We refer to this as coverage. The proposed approach is better than learning the dominant class in a tile as multiple soiling types occur within a tile commonly. It also has the advantage of dealing with coarse polygon annotation, which will cause the segmentation task. The proposed soiling coverage decoder is an order of magnitude faster than an equivalent segmentation decoder. We also integrated it into an object detection and semantic segmentation multi-task model using an asynchronous back-propagation algorithm. A portion of the dataset used will be released publicly as part of our WoodScape dataset to encourage further research.