优于平均：分割不确定性的空间感知聚合提升下游任务性能
Better than Average: Spatially-Aware Aggregation of Segmentation Uncertainty Improves Downstream Performance
对做分割不确定性量化的视觉研究者，本文系统比较了像素级不确定性到图像级分数的聚合策略，并给出利用空间结构的聚合器与元聚合器，可直接用于 OoD 检测与分割失败检测的下游流程。
原文依据 Global Average is the default choice, yet it does not account for spatial and structural features of segmentation uncertainty.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论不确定性量化（UQ）对于确保生物医学图像分析或自动驾驶等安全关键领域中自动图像分割的可靠性至关重要。在分割中，UQ 生成逐像素的不确定性分数，这些分数必须聚合为图像级分数，以用于分布外（OoD）检测或失败检测等下游任务。尽管聚合策略被常规使用，但其性质及对下游任务性能的影响尚未得到全面研究。全局平均是默认选择，但它并未考虑分割不确定性的空间与结构特征。基于块、基于类别和基于阈值的策略等替代方案虽然存在，但缺乏系统比较，导致报告不一致且最佳实践不明确。我们通过以下工作填补这一空白：(1) 形式化分析常见策略的性质、局限与陷阱；(2) 提出融入空间不确定性结构的新策略；(3) 在十个图像几何与结构各异的数据集上，对它们在 OoD 检测与失败检测上的性能进行基准评测。我们发现，利用空间结构的聚合器在所研究的两个下游任务中均取得更强性能。然而，单个聚合器的性能高度依赖数据集特征，因此我们 (4) 提出一种集成多个聚合器的元聚合器，可在不同数据集上稳健地工作。
英文原文摘要 1445 字符
Uncertainty Quantification (UQ) is crucial for ensuring the reliability of automated image segmentations in safety-critical domains like biomedical image analysis or autonomous driving. In segmentation, UQ generates pixel-wise uncertainty scores that must be aggregated into image-level scores for downstream tasks like Out-of-Distribution (OoD) or failure detection. Despite routine use of aggregation strategies, their properties and impact on downstream task performance have not yet been comprehensively studied. Global Average is the default choice, yet it does not account for spatial and structural features of segmentation uncertainty. Alternatives like patch-, class- and threshold-based strategies exist, but lack systematic comparison, leading to inconsistent reporting and unclear best practices. We address this gap by (1) formally analyzing properties, limitations, and pitfalls of common strategies; (2) proposing novel strategies that incorporate spatial uncertainty structure and (3) benchmarking their performance on OoD and failure detection across ten datasets that vary in image geometry and structure. We find that aggregators leveraging spatial structure yield stronger performance in both downstream tasks studied. However, the performance of individual aggregators depends heavily on dataset characteristics, so we (4) propose a meta-aggregator that integrates multiple aggregators and performs robustly across datasets.