Pith. sign in

REVIEW 4 cited by

CLIP-AD: A Language-Guided Staged Dual-Path Model for Zero-shot Anomaly Detection

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2311.00453 v2 pith:IDXNM32T submitted 2023-11-01 cs.CV

Xuhai Chen , Jiangning Zhang , Guanzhong Tian , Haoyang He , Wuhao Zhang , Yabiao Wang , Chengjie Wang , Yong Liu This is my paper
classification cs.CV
keywords modelanomalyfeaturestextzero-shotclip-addetectiondual-path
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

This paper considers zero-shot Anomaly Detection (AD), performing AD without reference images of the test objects. We propose a framework called CLIP-AD to leverage the zero-shot capabilities of the large vision-language model CLIP. Firstly, we reinterpret the text prompts design from a distributional perspective and propose a Representative Vector Selection (RVS) paradigm to obtain improved text features. Secondly, we note opposite predictions and irrelevant highlights in the direct computation of the anomaly maps. To address these issues, we introduce a Staged Dual-Path model (SDP) that leverages features from various levels and applies architecture and feature surgery. Lastly, delving deeply into the two phenomena, we point out that the image and text features are not aligned in the joint embedding space. Thus, we introduce a fine-tuning strategy by adding linear layers and construct an extended model SDP+, further enhancing the performance. Abundant experiments demonstrate the effectiveness of our approach, e.g., on MVTec-AD, SDP outperforms the SOTA WinCLIP by +4.2/+10.7 in segmentation metrics F1-max/PRO, while SDP+ achieves +8.3/+20.5 improvements.

Discussion (0). Sign in to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Keep the Needle, Prune the Haystack: Defect-Preserving Token Pruning for Efficient Zero-Shot Anomaly Detection

    cs.CV 2026-08 conditional novelty 7.0 of 10

    KeepAD prunes up to 80% of tokes in a frozen CLIP ViT for zero-shot anomaly detection, losing less than 2.7 AUROC points on average and achieving up to a 7.9x speedup over a strong CLIP baseline.

  2. Bridge Feature Matching and Cross-Modal Alignment with Mutual-filtering for Zero-shot Anomaly Detection

    cs.CV 2025-07 conditional novelty 6.0 of 10

    FiSeCLIP achieves state-of-the-art zero-shot anomaly detection by using a batch of test images as mutual references and filtering noisy features with text-guided masks, without any training.

  3. Closed form perturbative relativistic modifications to wave-packet dynamics in the quantum harmonic oscillator

    quant-ph 2026-03 unverdicted novelty 5.0 of 10

    Closed-form O(1/c²) relativistic corrections to QHO wave-packet widths, variances, and uncertainty products leave minimum-uncertainty saturation intact and become percent-level for 1–10 keV electron confinement.

  4. StackCLIP: Clustering-Driven Stacked Prompt in Zero-Shot Industrial Anomaly Detection

    cs.CV 2025-06 conditional novelty 4.0 of 10

    Stacking multiple category names in a CLIP text prompt, along with cluster-specific alignment layers, improves zero-shot industrial defect detection and localization.

Pith tools