ShiftSplit-AD: Separating Domain Shift from Defects in Foundation-Feature Visual Anomaly Detection

• URL: http://arxiv.org/abs/2608.27610v1
• Date: Thu, 27 Aug 2026 18:42:38 GMT
• Title: ShiftSplit-AD: Separating Domain Shift from Defects in Foundation-Feature Visual Anomaly Detection
• Abstract summary: We investigate whether structured decomposition of nearest-normal DINOv2 residuals can suppress shift-induced evidence.<n>ShiftSplit-AD decomposes the patch residual matrix into low-rank and row-sparse components.<n>Experiments expose a central trade-off rather than a universal separation.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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.

Related papers

• When does a spectral prior help graph learning? Connectivity-loss estimation under road-network disruptions [0.0]
  We estimate relative algebraic-connectivity loss after multi-edge deletion using graph neural networks (GNNs)<n>GCN, GraphSAGE, and edge-aware MPNN backbones are compared with analytical baselines.<n>Results characterize the spectral residual as a useful but domain-sensitive inductive bias for structural connectivity screening.
  arXiv  Detail & Related papers  (2026-09-10T07:12:22Z)
• Differentiable Interval Bottlenecks for Interpretable Anomaly Detection in Numerical Data [0.5213778368155993]
  DIFFINT is an autoencoder whose latent bottleneck is structured as a set of soft, axis-aligned interval memberships.<n>It is the only interpretable detector in the statistically-tied leading cluster of seven methods.
  arXiv  Detail & Related papers  (2026-09-03T14:05:27Z)
• Can You Trust Frozen Hematology Foundation Models under Acquisition Shift? [0.0]
  We audit 15 frozen encoders across four public single-cell acquisition domains.<n>In-domain linear-probe macro-F1 is saturated (0.98-0.997), yet cross-dataset macro-F1 drops 34-72%.<n>DinoBloom-L, the in-domain best, falls to 10th of 15 on the most-shifted target (MLL23), behind RedDino and several general and pathology encoders.
  arXiv  Detail & Related papers  (2026-08-25T20:55:57Z)
• Support Operation Factorization: Compositional Readout of Frozen Vision Encoders under Controlled Interventions [87.95757610455173]
  We introduce an injectively aligned leave-one-cell-out protocol over support x operation grids and SO-OPF.<n>We separate two questions that aggregate scores conflate: whether the carrier composes held-out bindings when the grid is known, and whether that grid can be recovered from flat cell labels.
  arXiv  Detail & Related papers  (2026-08-06T15:38:55Z)
• Post-Operative Glioma Segmentation via Loss Stabilization, Normalization and Subspace Attention [46.03321798937855]
  We show that the standard Generalized Dice Loss (GDL) is unstable under domain shift.<n>We propose a Subspace-Aware Class Attention (SACA) module that re-calibrates the bottleneck features and raises Enhancing Tumor (ET) sensitivity by 8%.
  arXiv  Detail & Related papers  (2026-07-23T09:59:30Z)
• Domain Shift in Echocardiography: Interpretable Quantification and Prediction of Cross-Dataset Left Ventricular Segmentation [0.5599792629509228]
  Cross-dataset generalisation remains a major barrier to clinical deployment of echocardiographic left ventricular segmentation.<n>We examined whether transfer degradation could be estimated before deployment using ultrasound descriptors, VAE latent features, and segmentation-derived latent features.
  arXiv  Detail & Related papers  (2026-07-22T00:48:57Z)
• UniVAD v2: Unified Visual Anomaly Detection via Support-Conditioned Boundary Construction [40.70728779828606]
  We propose UniVAD v2, a two-sided support construction spanning unified visual anomaly detection.<n>Built on the component-patch divide-and-conditioned framework of UniVAD, UniVAD v2 strengthens the normal side with an Optimal Transport Modeling module (OTRM)<n>On the abnormal side, a Few-Shot Abnormal Reference module (FAR) converts abnormal references into rejection-side evidence for boundary adjustment.
  arXiv  Detail & Related papers  (2026-06-29T02:38:51Z)
• Unsupervised Domain Shift Detection with Interpretable Subspace Attribution [48.553902624178995]
  Domain shifts are subtle differences in the probability distributions of datasets.<n>We identify these shifts using an algorithm designed to detect localised density anomalies in high-dimensional feature spaces.<n>We trace the domain shift to a small set of features, making the shift interpretable.
  arXiv  Detail & Related papers  (2026-05-15T12:58:00Z)
• D3R-Net: Dual-Domain Denoising Reconstruction Network for Robust Industrial Anomaly Detection [0.0]
  Unsupervised anomaly detection (UAD) is a key ingredient of automated visual inspection in modern manufacturing.<n>We introduce D3R-Net, a Dual-Domain Denoising Reconstruction framework that couples a self-supervised 'healing' task with frequency-aware regularization.<n>In addition to the spatial mean squared error, we employ a Fast Fourier Transform (FFT) magnitude loss that encourages consistency in the frequency domain.
  arXiv  Detail & Related papers  (2026-01-27T23:21:59Z)
• Hide-and-Seek Attribution: Weakly Supervised Segmentation of Vertebral Metastases in CT [68.09387763135236]
  We introduce a weakly supervised method trained solely on vertebra-level healthy/malignant labels, without any lesion masks.<n>We achieve strong blastic/lytic performance despite no mask supervision.
  arXiv  Detail & Related papers  (2025-12-07T14:03:28Z)
• I Detect What I Don't Know: Incremental Anomaly Learning with Stochastic Weight Averaging-Gaussian for Oracle-Free Medical Imaging [2.384534878752428]
  We introduce an unsupervised, oracle-free framework that incrementally expands a trusted set of normal samples without any anomaly labels.<n>A frozen pretrained vision backbone is augmented with tiny convolutional adapters, ensuring rapid domain adaptation with negligible computational overhead.<n>On COVID-CXR, ROC-AUC improves from 0.9489 to 0.9982; on Pneumonia CXR, ROC-AUC rises from 0.6834 to 0.8968; and on Brain MRI ND-5, ROC-AUC increases from 0.6041 to 0.7269.
  arXiv  Detail & Related papers  (2025-11-05T23:28:14Z)
• MSFlow: Multi-Scale Flow-based Framework for Unsupervised Anomaly Detection [124.52227588930543]
  Unsupervised anomaly detection (UAD) attracts a lot of research interest and drives widespread applications. An inconspicuous yet powerful statistics model, the normalizing flows, is appropriate for anomaly detection and localization in an unsupervised fashion. We propose a novel Multi-Scale Flow-based framework dubbed MSFlow composed of asymmetrical parallel flows followed by a fusion flow. Our MSFlow achieves a new state-of-the-art with a detection AUORC score of up to 99.7%, localization AUCROC score of 98.8%, and PRO score of 97.1%.
  arXiv  Detail & Related papers  (2023-08-29T13:38:35Z)
• ADTR: Anomaly Detection Transformer with Feature Reconstruction [40.68590890351697]
  Anomaly detection with only prior knowledge from normal samples attracts more attention. Existing CNN-based pixel reconstruction approaches suffer from two concerns. We propose Anomaly Detection TRansformer (ADTR) to apply a transformer to reconstruct pre-trained features.
  arXiv  Detail & Related papers  (2022-09-05T08:01:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.