DIVE improves zero-shot anomaly detection under limited auxiliary anomaly priors via shallow-and-deep text embedding injection and disentanglement, outperforming baselines by up to 28.5% on classification and 47.0% on segmentation across twelve datasets.
Anovl: Adapting vision-language models for unified zero-shot anomaly localization.arXiv preprint arXiv:2308.15939, 2(5)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2representative citing papers
A training-free method fits PCA to DINOv2 features from few normal images and detects anomalies via reconstruction residual, reaching SOTA one-shot AUROC of 97.1% image-level on MVTec-AD and 93.2% on VisA.
citing papers explorer
-
Robust Zero-shot Anomaly Detection under Limited Auxiliary Anomaly Priors
DIVE improves zero-shot anomaly detection under limited auxiliary anomaly priors via shallow-and-deep text embedding injection and disentanglement, outperforming baselines by up to 28.5% on classification and 47.0% on segmentation across twelve datasets.
-
SubspaceAD: Training-Free Few-Shot Anomaly Detection via Subspace Modeling
A training-free method fits PCA to DINOv2 features from few normal images and detects anomalies via reconstruction residual, reaching SOTA one-shot AUROC of 97.1% image-level on MVTec-AD and 93.2% on VisA.