ShiftSplit-AD：在基础特征视觉异常检测中分离域偏移与缺陷
ShiftSplit-AD: Separating Domain Shift from Defects in Foundation-Feature Visual Anomaly Detection
浏览论文内容
中文总结 AI 辅助
ShiftSplit-AD通过分解DINOv2残差分离域偏移与缺陷，在AeBAD-S上提升了图像AUROC和AUPRC，但存在缺陷结构保留的权衡问题。
中文摘要 AI 辅助
基于冻结基础模型特征的视觉异常检测器通常会对测试块与正常特征记忆库的距离进行评分，但良性采集变化也会增大这些距离，混淆域变化与缺陷。本研究探究对最近正常的DINOv2残差进行结构化分解是否能抑制偏移诱导的证据，同时保留未见过的缺陷。ShiftSplit-AD将块残差矩阵分解为低秩和行稀疏分量，并对稀疏分量进行评分，可选低秩/稀疏融合。实验揭示了核心权衡而非通用分离：真实缺陷可包含相关低维结构，因此过滤广泛的残差活动也可能去除缺陷信息。在AeBAD-S数据集上，使用经Bottle类别开发后固定的设置，仅稀疏评分使图像AUROC从0.6780提升至0.7294，AUPRC从0.8052提升至0.8465；改进的配对自助法95%置信区间分别为[0.0238, 0.0808]和[0.0170, 0.0650]。但仅稀疏评分在四个保留的MVTec类别上将平均纯净AUROC从0.9890降至0.9133，并降低了Bottle的定位性能。这些发现表明，当域偏移严重污染异常证据时，残差分解可提供帮助，但保留缺陷结构仍是限制问题。
英文摘要
Visual anomaly detectors based on frozen foundation-model features commonly score distances from test patches to a memory of normal features. Benign acquisition changes can also enlarge these distances, confounding domain variation with defects. We investigate whether structured decomposition of nearest-normal DINOv2 residuals can suppress shift-induced evidence while retaining unseen defects. ShiftSplit-AD decomposes the patch residual matrix into low-rank and row-sparse components and scores the sparse component, with an optional low-rank/sparse fusion. The experiments expose a central trade-off rather than a universal separation: genuine defects can contain correlated, low-dimensional structure, so filtering broad residual activity may also remove defect information. On AeBAD-S, using settings fixed after Bottle development, sparse-only scoring improves image AUROC from 0.6780 to 0.7294 and AUPRC from 0.8052 to 0.8465. Paired bootstrap 95% intervals for the improvements are [0.0238, 0.0808] and [0.0170, 0.0650], respectively. However, sparse-only scoring reduces mean clean AUROC from 0.9890 to 0.9133 on four held-out MVTec categories and degrades Bottle localization. These findings show that residual decomposition can help when domain shift strongly contaminates anomaly evidence, but preserving defect structure remains the limiting problem.
发表机构
- University of Moratuwa（莫拉图瓦大学）
机构由 AI 辅助整理，请以论文原文为准。
