U-CE: Uncertainty-aware Cross-Entropy for Semantic Segmentation

Steven Landgraf, Markus Hillemann, Kira Wursthorn, Markus Ulrich

Abstract:

Deep neural networks have shown exceptional performance in various tasks, but their lack of robustness, reliability, and tendency to be overconfident pose challenges for their deployment in safety-critical applications like autonomous driving. In this regard, quantifying the uncertainty inherent to a model’s prediction is a promising endeavour to address these shortcomings. In this work, we
present a novel Uncertainty-aware Cross-Entropy loss (UCE) that incorporates dynamic predictive uncertainties into
the training process by pixel-wise weighting of the wellknown cross-entropy loss (CE). Through extensive experimentation, we demonstrate the superiority of U-CE over regular CE training on two benchmark datasets, Cityscapes
and ACDC, using two common backbone architectures, ResNet-18 and ResNet-101. With U-CE, we manage to
train models that not only improve their segmentation performance but also provide meaningful uncertainties after
training. Consequently, we contribute to the development of more robust and reliable segmentation models, ultimately
advancing the state-of-the-art in safety-critical applications and beyond.