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. In particular, we observe superior performance in terms of OoD segmentation to comparable baselines on the SegmentMeIfYouCan benchmark, clearly outperforming methods which are similarly flexible to implement.

翻译：近年来，深度神经网络在语义分割任务中定义了最先进水平，其预测局限于预设的语义类别集合。这类模型将被部署于自动驾驶等应用中，然而其分类受限的表达能力与开放世界场景存在根本矛盾。因此，检测并分割超出预设语义空间的目标（即分布外目标）具有极高研究价值。由于软最大熵或贝叶斯模型等不确定性估计方法对错误预测敏感，这些方法自然成为分布外检测的基准。本文提出一种通过逐像素损失梯度获取不确定性分数的方法，该分数可在推理阶段高效计算。该方法适用于大多数模型，实现简便，无需额外训练或辅助数据，可直接应用于预训练分割模型。实验表明，本方法能有效识别错误像素分类并评估预测质量。特别地，在SegmentMeIfYouCan基准上，本方法在分布外分割任务中展现出优于可比基线方法的性能，显著超过同类灵活实现方法。