The fusion of raw features from multiple sensors on an autonomous vehicle to create a Bird's Eye View (BEV) representation is crucial for planning and control systems. There is growing interest in using deep learning models for BEV semantic segmentation. Anticipating segmentation errors and improving the explainability of DNNs is essential for autonomous driving, yet it is under-studied. This paper introduces a benchmark for predictive uncertainty quantification in BEV segmentation. The benchmark assesses various approaches across three popular datasets using two representative backbones and focuses on the effectiveness of predicted uncertainty in identifying misclassified and out-of-distribution (OOD) pixels, as well as calibration. Empirical findings highlight the challenges in uncertainty quantification. Our results find that evidential deep learning based approaches show the most promise by efficiently quantifying aleatoric and epistemic uncertainty. We propose the Uncertainty-Focal-Cross-Entropy (UFCE) loss, designed for highly imbalanced data, which consistently improves the segmentation quality and calibration. Additionally, we introduce a vacuity-scaled regularization term that enhances the model's focus on high uncertainty pixels, improving epistemic uncertainty quantification.

Video games are playing a big part in helping people cope during the coronavirus pandemic. Since earlier this spring with the onset of stay-at-home orders meant to stem the spread of COVID-19, more Americans have pressed play on video games. For some, games are an entertaining way to pass the time not spent on other pursuits. Others use them to stay connected with friends they used to see in person – and to bond with family members. Jennifer Fidler, 47, and her husband of Portland, Oregon, have been playing "Animal Crossing: New Horizons" with her two middle school-aged daughters since the pandemic led to school closings.