用对比双高斯过程实现弱监督异常分割，无需像素标注
CDGP: Contrastive Dual Gaussian Processes for Weakly Supervised Anomaly Segmentation
- 构建正负样本对比的双高斯过程模型，捕捉密集特征分布
- 在MVTec AD2上所有定位指标排名第一，性能超越现有方法
- 仅需正常样本训练，适合无像素标签的工业质检场景
工业视觉检测需判断产品是否缺陷并定位缺陷区域，但像素级标注成本高昂。现有异常分割方法多仅基于正常图像学习，通过偏离正常程度评分。然而真实缺陷与异常但正常的区域都可能产生高分，难以区分。本文提出对比双高斯过程（CDGP），在密集特征片段上建模正常与异常诱导变量的预测分布。其后验主导统计量以联合预测不确定性标准化预测均值差异，同时提供空间证据和图像级置信度。该证据与层次化正常重建残差互补，实现精细定位。所有校准仅使用训练数据，无需人工像素标注或测试时调优。在MVTec AD2、KSDD2和VisA三个数据集上，CDGP在所有MVTec AD2定位指标中排名第一，在KSDD2和VisA上亦为第一或具有竞争力。因子分解与匹配线性头用于界定线性核高斯过程形式的贡献与范围。
原文摘要 · Abstract (English)
Industrial visual inspection must both decide whether a product is defective and localize the defect, yet pixel-level masks are costly to collect at scale. Most anomaly-segmentation methods learn only from defect-free images and score deviations from normality. A true defect and an unusual-but-normal region, however, can both deviate substantially and receive similarly high scores. We propose Contrastive Dual Gaussian Processes (CDGP), a weakly supervised framework that models normal and anomaly inducing-variable predictive distributions over dense tokens. Its posterior-dominance statistic standardizes their predictive-mean difference by the joint predictive uncertainty, providing both spatial evidence and image-level confidence. This evidence complements hierarchical normal-reconstruction residuals for fine localization. All calibration uses training data only, without human pixel annotations or test-time fitting. Across MVTec AD~2, KSDD2, and VisA, CDGP ranks first among the evaluated methods on all MVTec AD~2 localization metrics and is first-place or competitive on KSDD2 and VisA. Factorized and matched linear-head controls delimit the contribution and scope of the linear-kernel Gaussian process (GP) formulation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。