DAUPNet：面向跨域少样本语义分割的不确定性感知原型判别方法
DAUPNet: Domain-Aware Uncertainty Modeling for Reliable Prototype Discrimination in Cross-Domain Few-Shot Semantic Segmentation
AI 导读
DAUPNet 将跨域少样本语义分割（CD-FSS）中的原型匹配重构为不确定性感知的原型判别框架，通过协调支持-查询层次特征、概率化建模前景与背景原型，并用估计的不确定性调节对比优化。在四个标准目标域上，该方法 1-shot 和 5-shot 设置分别达到 72.6% 和 76.7% 的平均 mIoU，在两个医学领域也有显著提升。
正文 · 原文
Abstract:Cross-domain few-shot semantic segmentation (CD-FSS) has predominantly been formulated as learning domain-invariant representations or improving support-query correspondence. Nevertheless, large domain shifts still make prototype matching unreliable: inconsistent hierarchical responses corrupt the support representation, deterministic prototypes cannot express boundary and appearance ambiguity, and treating prototypes with different reliability equally during optimization weakens foreground-background separation. We therefore propose DAUPNet, a unified framework that reformulates cross-domain prototype matching as uncertainty-aware prototype discrimination. DAUPNet first harmonizes hierarchical support-query features to provide stable evidence, then represents foreground and background prototypes probabilistically, and finally uses their estimated uncertainty to regulate contrastive optimization. On four standard target domains, DAUPNet achieves 72.6% and 76.7% average mIoU in the 1-shot and 5-shot settings, respectively, including substantial gains on the two medical domains. These results demonstrate that modeling prototype uncertainty and incorporating it into optimization provides a robust and interpretable approach to CD-FSS under severe domain shift. The code is available at this https URL
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2607.16308 [cs.CV] |
| (or arXiv:2607.16308v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2607.16308 arXiv-issued DOI via DataCite |
Submission history
From: Lei Yuan [view email]
[v1]
Tue, 14 Jul 2026 14:48:54 UTC (15,480 KB)
[v2]
Fri, 30 Oct 2026 12:57:45 UTC (18,052 KB)
来源：arXiv cs.AI · arxiv.org