Pixel-Wise Gradient Uncertainty for Convolutional Neural Networks Applied to Out-of-Distribution Segmentation

Kira Maag, Tobias Riedlinger

2024

Abstract

In recent years, deep neural networks have defined the state-of-the-art in semantic segmentation where their predictions are constrained to a predefined set of semantic classes. They are to be deployed in applications such as automated driving, although their categorically confined expressive power runs contrary to such open world scenarios. Thus, the detection and segmentation of objects from outside their predefined semantic space, i.e., out-of-distribution (OoD) objects, is of highest interest. Since uncertainty estimation methods like softmax entropy or Bayesian models are sensitive to erroneous predictions, these methods are a natural baseline for OoD detection. Here, we present a method for obtaining uncertainty scores from pixel-wise loss gradients which can be computed efficiently during inference. Our approach is simple to implement for a large class of models, does not require any additional training or auxiliary data and can be readily used on pre-trained segmentation models. Our experiments show the ability of our method to identify wrong pixel classifications and to estimate prediction quality at negligible computational overhead. In particular, we observe superior performance in terms of OoD segmentation to comparable baselines on the SegmentMeIfYouCan benchmark, clearly outperforming other methods.

Paper Citation

in Harvard Style

Maag K. and Riedlinger T. (2024). Pixel-Wise Gradient Uncertainty for Convolutional Neural Networks Applied to Out-of-Distribution Segmentation. In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP; ISBN 978-989-758-679-8, SciTePress, pages 112-122. DOI: 10.5220/0012353300003660

in Bibtex Style

@conference{visapp24,
author={Kira Maag and Tobias Riedlinger},
title={Pixel-Wise Gradient Uncertainty for Convolutional Neural Networks Applied to Out-of-Distribution Segmentation},
booktitle={Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP},
year={2024},
pages={112-122},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012353300003660},
isbn={978-989-758-679-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP
TI - Pixel-Wise Gradient Uncertainty for Convolutional Neural Networks Applied to Out-of-Distribution Segmentation
SN - 978-989-758-679-8
AU - Maag K.
AU - Riedlinger T.
PY - 2024
SP - 112
EP - 122
DO - 10.5220/0012353300003660
PB - SciTePress