Pith. sign in

REVIEW 3 major objections 5 minor 46 references

One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection

T0 review · 3 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read The paper's central claim is that few-shot anomaly detection can be sharpened by first generating a personalized normal version of each query image with a customized diffusion model and then scoring the query against that normal twin, a…

desk verdict Promising idea with a load-bearing flaw in the prompt-selection objective (Eq. 2) that the authors must fix before the one-to-normal claim is credible. read the letter →

arxiv 2502.01201 v1 pith:FKB55IVN submitted 2025-02-03 cs.CV

Yiyue Li , Shaoting Zhang , Kang Li , Qicheng Lao This is my paper · ORCID
classification cs.CV
keywords few-shotanomalydetectionpersonalizationone-to-normaltransformationdiffusionmodelsDreamboothcustomizationtripletcontrastiveinferenceCLIPanomaly-freesamplegeneration
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Few-shot anomaly detection usually compares a query image directly against a few stored normal reference images, a step that is easily confused by benign differences in pose, texture, or lighting. This paper proposes instead to first transform the query image itself into a personalized 'normal' version using a diffusion model fine-tuned on the few normal examples, so that anomalies are erased while normal structure is kept. The query is then scored in three complementary ways: against its personalized twin, against a pool of anomaly-free samples (real and generated), and against text prompts. The authors report that this triplet contrastive scheme outperforms existing few-shot anomaly detection methods across eleven datasets in industrial, medical, and semantic domains, with the largest gains in the 8-shot setting, and that the generated anomaly-free images can be dropped into other anomaly detectors to improve them.

What carries the argument

The load-bearing mechanism is the one-to-normal transformation: a query image $x_q$ is partially noised to $x_{q,t}$ and denoised by the anomaly-free customized diffusion model $D_\theta$ under the text prompt $c_q$ chosen to minimize SSIM loss against the query, producing a personalized image $\hat{x}_q = D_\theta(\sqrt{\alpha_t} x_q + \sqrt{1-\alpha_t}\epsilon, c_q, t)$. This is the operation that maps the query onto the normal manifold, erasing anomalies while preserving identity. The second mechanism is the triplet contrastive anomaly inference, which pools three scores—$S_P$ from multi-level cosine dissimilarity between query and personalized image, $S_N$ from the nearest memory-bank feature among anomaly-free samples, and $S_{\text{text}}$ from softmax over CLIP text features—into the final score. The customization itself rests on a Dreambooth fine-tune of Stable Diffusion with a class-specific prior-preservation loss so that the model generates normal-looking samples without language drift.

What would settle it

Take a defect-free but visually atypical object (e.g., a normal image with unusual pose, lighting, or background) as query, run the one-to-normal personalization, and measure whether the pixel-level difference between query and personalized image is concentrated in genuinely anomalous regions; if normal regions shift as much as anomalous ones, or if a deliberately injected defect survives the transformation, the central premise is refuted. Concretely, on MVTec-AD with known ground-truth masks, compute the AUROC of a defect-localization score defined by the per-pixel difference between $x_q$ and $\hat{x}_q$; if it is no better than chance on categories with high pose variation, the personalization is not selectively removing anomalies.

Watch

Extended reading notes

Core claim

The central claim is that precise and stable few-shot anomaly detection can be achieved by 'anomaly personalization': rather than matching the query to a few reference images, one matches it to its own one-to-normal transformation. An anomaly-free customized model—a diffusion model fine-tuned through Dreambooth on two to eight normal images per object, with data augmentation and prior-preservation loss—learns the normal distribution. At test time the query image is partially noised (t-step ratio 0.3) and denoised under a selected normal-state text prompt, yielding a personalized image that keeps normal regions and repairs defective ones. The final anomaly score combines three contrastive terms: multi-level feature similarity between query and personalized image ($S_P$), similarity between query and a memory bank of anomaly-free samples including generated ones ($S_N$), and CLIP text-prompt classification ($S_{\text{text}}$), summed as $A_{\text{score}} = S_P + \alpha S_N + \beta S_{\text{text}}$. The paper reports that this beats previous few-shot AD methods on MVTec, VisA, KSDD, AFID, ELPV, OCT2017, BrainMRI, HeadCT, RESC, MNIST, and CIFAR-10, with AUROC gains up to about four points over the second-best method in 8-shot settings, and that generated normal images improve other detectors such as PatchCore, RegAD, WinCLIP, and InCTRL when added to their reference sets.

Load-bearing premise

The whole method assumes that partially noising a query image at t-step 0.3 and denoising it with a diffusion model trained on two to eight normal images will erase the anomaly while preserving the object's normal appearance; if that transformation instead keeps the defect or destroys normal structure, the personalized comparison becomes uninformative or worse than direct reference matching.

Editorial extensions

If this is right

  • If the central claim holds, few-shot anomaly detection no longer needs to rely on direct query-to-reference feature matching; the query's own normal twin provides a finer-grained comparison.
  • The generated anomaly-free samples form a reusable augmentation: adding them to the reference pools of existing detectors improves their AUROC, with reported gains of up to about ten points for some baselines.
  • The method transfers across industrial, medical, and semantic domains, suggesting that the personalization approach is not tied to object categories seen during pretraining.
  • Because the customized model is frozen after Dreambooth fine-tuning and only the inference-time noise/denoise step is new, the pipeline can be grafted onto existing CLIP-based few-shot AD systems at a modest inference cost of a few hundred milliseconds per image.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The one-to-normal transformation is essentially a per-query data augmentation; a testable extension is to use the personalized image as a 'normal twin' for anomaly localization via pixel-wise residual maps, which the paper only shows qualitatively.
  • Because the t-step ratio (0.3) is fixed across all datasets, one can infer that a per-domain or per-category t-step schedule could further improve results; the paper does not explore this.
  • The method's reliance on a few normal images per category means it cannot go zero-shot; an implicit consequence is that the same personalization machinery could be applied to a generic normal prior for unseen categories, but that would require the customized model to generalize across objects.
  • The triplet score weighting $\alpha=1, \beta=0.5$ is fixed; an inference is that the text branch matters less in medical domains where CLIP has weak domain knowledge, so adaptive weighting could shift the balance.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes a few-shot anomaly detection method built on a DreamBooth-style customized diffusion model. Given a small set of normal reference images, the method fine-tunes a diffusion model to model the normal distribution, then 'personalizes' each query image by reconstructing it at a fixed noise level under a text prompt selected from a set of normal-state prompts. The resulting personalized image is compared with the query, together with an anomaly-free sample memory bank and text-prompt scores, through a triplet contrastive inference scheme whose weighted combination gives the final anomaly score. The authors report experiments on 11 datasets in industrial, medical, and semantic domains, claim consistent AUROC/AUPRC improvements over several few-shot AD baselines, and show that the generated anomaly-free samples can improve other AD methods when used as augmentation.

Significance. The empirical scope is a genuine strength: 11 datasets across three domains, three few-shot settings, five runs, ablations of each inference branch and of the prompt design, plus transfer experiments with generated samples used as training data for PaDiM, PatchCore, WinCLIP, and InCTRL. If the one-to-normal premise were established, the method would be a valuable contribution to few-shot anomaly detection. However, as written, the central mechanism is not established: the prompt-selection objective in Eq. (2) is internally inconsistent, and the paper provides no quantitative verification that the personalized images actually remove anomalies while preserving normal structure. The empirical claims therefore do not yet validate the proposed mechanism, and the contribution is conditional on a corrected and verified formulation.

major comments (3)
  1. [Sec. 3.3, Eq. (2)] The prompt-selection objective is stated as cq = argmin_ci L(xq, Dθ(xq,t, ci, t)) with L identified as SSIM. An argmin over SSIM selects the reconstruction least similar to the query, which is the opposite of the 'most closely resembles the normal state' goal stated in the same paragraph. If L is instead intended as a dissimilarity such as 1−SSIM, the objective still only compares with xq, not with any normal reference image or with the learned normal manifold Pnormal. For an anomalous query, the prompt that maximizes similarity to xq is the one that best preserves the anomaly's appearance. Since Table 5 shows that the prompt selection changes results (e.g., MVTec-AD 95.6 to 96.2, OCT2017 98.7 to 99.3), the ambiguity is consequential. Please correct the sign and re-define the selection criterion relative to normal references or the normal-state distribution, and add evidence that the selected prompt does not preserve anomalies.
  2. [Sec. 3.3, Eq. (3) and t hyperparameter] The one-to-normal transformation uses t = 0.3 with a noise schedule in which the generated image is strongly conditioned by xq. At this noise level, the text prompt has limited influence and the reconstruction may retain the anomaly; conversely, if the schedule is more aggressive it may erase normal structure. No sensitivity analysis or quantitative verification that Dθ maps anomalous queries onto Pnormal is provided. Please report AUROC as a function of t and show, for example, that personalized images of anomalous queries are closer to normal reference features than the original queries are, or provide an alternative validation of the core premise.
  3. [Sec. 4.1 and Eq. (7)] The final score combines SP, SN, and Stext with weights α = 1 and β = 0.5, and the personalization strength t = 0.3 is chosen globally. The manuscript does not describe a validation protocol (e.g., a held-out split) for these choices, and the same 11 datasets are used for both selection and reporting. Because the branch ablation in Table 4 shows that the relative contributions vary strongly by dataset (e.g., Stext alone gives 45.3 on OCT2017 but 96.3 on KSDD), the fixed weights may mask overfitting or underfitting to particular domains. Please report sensitivity to α, β, and t and clarify the selection protocol.
minor comments (5)
  1. [Sec. 3.2, Eq. (1)] Eq. (1) has an unmatched closing parenthesis in ||Dθ(xnormal,t, c) − x0||2, and the notation xnormal,t is not defined before it is used in Eq. (2); please define the forward noising process once and use it consistently.
  2. [Sec. 4.2, paragraph after Table 1] The sentence reporting 8-shot improvements lists six datasets (KSDD, ELPV, AFID, KSDD, VisA, MVTec) but gives five percentage values and duplicates KSDD; please reconcile the text with the values in Table 1.
  3. [Table 1] The AFID row in the 2-shot block is missing a closing parenthesis: '(78.3±1.7, 53.6±1.2' should be '(78.3±1.7, 53.6±1.2)'.
  4. [Sec. 4.3, Inference time] The inference-time paragraph states that the proposed method is '+200-300ms per query image' higher than WinCLIP (389ms) and InCTRL (276ms), then says that the single-prompt configuration requires 326ms, which is lower than WinCLIP; please clarify which configuration the +200-300ms figure refers to.
  5. [Appendix A.1] The main text says three prompts were used for one-to-normal personalization, while the appendix describes experiments with 1, 3, 5, and 10 prompts and says all ten gave the best performance; please state the exact number and contents of prompts used in the reported main results.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the central claims are empirical and the score composition is not fitted to the reported predictions by construction.

full rationale

The paper's pipeline (Section 3) trains a Dreambooth-style diffusion model on few-shot normal images (Eq. 1), personalizes each query through a diffusion reconstruction (Eqs. 2-3), and combines three CLIP-based anomaly scores with fixed weights (Eq. 7). None of these equations defines an output as the value of a fitted parameter or as a re-labeling of the target variable. alpha=1 and beta=0.5 are stated as dataset-independent constants, and t=0.3 is a hyperparameter; even if tuned on the benchmark, this is a fitting risk, not a circular reduction. The one self-citation ([21]) appears only as a domain citation in the introduction and is not load-bearing. The most serious validity concern is Eq. (2): the prompt cq is selected by minimizing L(xq, Dtheta(...)) with L stated as SSIM, so the claim that this selects the image that 'most closely resembles the normal state' is not supported by the equation (argmin over SSIM is least similar, and the objective has no normal reference image). This is an unsupported-mechanism/correctness issue rather than a circularity: the personalized image is not by construction identical to the normal manifold, and the reported AUROC improvements are empirical outcomes that could fail if the mechanism is flawed. Overall, the derivation chain does not reduce any claimed prediction to its inputs by construction.

Assumptions & free parameters 5 free parameters · 3 assumptions · 0 invented entities

The method rests on the assumption that a diffusion model fine-tuned on a few normal images can faithfully project an anomalous query onto the normal manifold. The final score uses hand-picked weights and a fixed noise schedule that are not derived from first principles. No new physical or mathematical entities are introduced.

free parameters (5)
  • t-step ratio = 0.3
    Controls the strength of one-to-normal personalization; fixed across datasets; likely tuned to achieve good performance.
  • alpha (weight for anomaly-free sample score) = 1
    Weight for the anomaly-free sample comparison score in the final Ascore.
  • beta (weight for text prompt score) = 0.5
    Weight for the text prompt comparison score in the final Ascore.
  • memory bank size = 30
    Number of generated anomaly-free samples stored for the SN comparison.
  • number of text prompts for personalization = 10
    Prompt count selected after comparing 1, 3, 5, and 10 prompts; 10 gave the highest AUROC (Appendix A.1).
assumptions (3)
  • domain assumption Diffusion personalization removes anomalies while preserving normal structure.
    Sec. 3.3 assumes the customized model transforms anomalous regions to normal while preserving normal appearance.
  • domain assumption Generated anomaly-free samples represent the normal distribution.
    Sec. 3.4 relies on samples from the customized model as proxies for normal images in the memory bank.
  • domain assumption CLIP features capture anomaly-relevant differences.
    Used in Eqs. (4)-(6); CLIP is pre-trained and frozen, and the paper assumes its multi-level features are sensitive to anomalies.

how reviews work

0 comments
Cite this review

Pith. "Pith review of One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection." pith.science (2026). https://pith.science/paper/FKB55IVN

@misc{pith2026250201201,
  author       = {Pith},
  title        = {Pith review of: One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FKB55IVN}},
  note         = {Machine review of arXiv:2502.01201}
}
read the original abstract

Traditional Anomaly Detection (AD) methods have predominantly relied on unsupervised learning from extensive normal data. Recent AD methods have evolved with the advent of large pre-trained vision-language models, enhancing few-shot anomaly detection capabilities. However, these latest AD methods still exhibit limitations in accuracy improvement. One contributing factor is their direct comparison of a query image's features with those of few-shot normal images. This direct comparison often leads to a loss of precision and complicates the extension of these techniques to more complex domains--an area that remains underexplored in a more refined and comprehensive manner. To address these limitations, we introduce the anomaly personalization method, which performs a personalized one-to-normal transformation of query images using an anomaly-free customized generation model, ensuring close alignment with the normal manifold. Moreover, to further enhance the stability and robustness of prediction results, we propose a triplet contrastive anomaly inference strategy, which incorporates a comprehensive comparison between the query and generated anomaly-free data pool and prompt information. Extensive evaluations across eleven datasets in three domains demonstrate our model's effectiveness compared to the latest AD methods. Additionally, our method has been proven to transfer flexibly to other AD methods, with the generated image data effectively improving the performance of other AD methods.

Figures

Figures reproduced from arXiv: 2502.01201 by the authors.

Figure 1
Figure 1. Overview of our proposed anomaly personalization approach. First, we use few-shot normal [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Visualization of representative results for pixel-level anomaly localization of our proposed [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Visualizations of anomaly personalization. The red box of the query image indicates [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (13 more)
Figure 4
Figure 4. Figure 4: The effectiveness of our generated anomaly-free samples for other AD methods. The red [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Lists of state and template level prompts employed in this paper to construct text features. [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: The anomaly map localization results of our proposed method for subclass chewinggum. [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 7
Figure 7. Figure 7: The anomaly map localization results of our proposed method for subclass fryum. The [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]
Figure 8
Figure 8. Figure 8: The anomaly map localization results of our proposed method for subclass candle. The [PITH_FULL_IMAGE:figures/full_fig_p015_8.png]
Figure 9
Figure 9. Figure 9: The anomaly map localization results of our proposed method for subclass pcb2. The [PITH_FULL_IMAGE:figures/full_fig_p015_9.png]
Figure 10
Figure 10. Figure 10: The anomaly map localization results of our proposed method for subclass grid. The [PITH_FULL_IMAGE:figures/full_fig_p015_10.png]
Figure 11
Figure 11. Figure 11: The anomaly map localization results of our proposed method for subclass hazelnut. The [PITH_FULL_IMAGE:figures/full_fig_p015_11.png]
Figure 12
Figure 12. Figure 12: The anomaly map localization results of our proposed method for subclass metal_nut. The [PITH_FULL_IMAGE:figures/full_fig_p016_12.png]
Figure 13
Figure 13. Figure 13: The anomaly map localization results of our proposed method for subclass tile. The [PITH_FULL_IMAGE:figures/full_fig_p016_13.png]
Figure 14
Figure 14. Figure 14: The anomaly map localization results of our proposed method for subclass BrainCT. The [PITH_FULL_IMAGE:figures/full_fig_p016_14.png]
Figure 15
Figure 15. Figure 15: The anomaly map localization results of our proposed method for subclass HeadCT. The [PITH_FULL_IMAGE:figures/full_fig_p016_15.png]
Figure 16
Figure 16. Figure 16: The anomaly map localization results of our proposed method for subclass OCT. The [PITH_FULL_IMAGE:figures/full_fig_p017_16.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 30 canonical work pages

  1. [1]

    Diffusion-based data augmentation for skin disease classification: Impact across original medical datasets to fully synthetic images

    Mohamed Akrout, Bálint Gyepesi, Péter Holló, Adrienn Poór, Blága Kincs ˝o, Stephen Solis, Katrina Cirone, Jeremy Kawahara, Dekker Slade, Latif Abid, et al. Diffusion-based data augmentation for skin disease classification: Impact across original medical datasets to fully synthetic images. In International Conference on Medical Image Computing and Computer...

    work page 2023

  2. [2]

    Mvtec ad–a comprehensive real-world dataset for unsupervised anomaly detection

    Paul Bergmann, Michael Fauser, David Sattlegger, and Carsten Steger. Mvtec ad–a comprehensive real-world dataset for unsupervised anomaly detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 9592–9600, 2019

    work page 2019

  3. [3]

    Anomaly detection under distribution shift

    Tri Cao, Jiawen Zhu, and Guansong Pang. Anomaly detection under distribution shift. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 6511–6523, 2023

    2023

  4. [4]

    Subject-driven text-to-image generation via apprenticeship learning

    Wenhu Chen, Hexiang Hu, Yandong Li, Nataniel Ruiz, Xuhui Jia, Ming-Wei Chang, and William W Cohen. Subject-driven text-to-image generation via apprenticeship learning. Advances in Neural Information Processing Systems, 36, 2024

    work page 2024

  5. [5]

    A zero-/few-shot anomaly classification and segmentation method for cvpr 2023 vand workshop challenge tracks 1&2: 1st place on zero-shot ad and 4th place on few-shot ad

    Xuhai Chen, Yue Han, and Jiangning Zhang. A zero-/few-shot anomaly classification and segmentation method for cvpr 2023 vand workshop challenge tracks 1&2: 1st place on zero-shot ad and 4th place on few-shot ad. arXiv preprint arXiv:2305.17382, 2023

    arXiv 2023

  6. [6]

    Padim: a patch distribution modeling framework for anomaly detection and localization

    Thomas Defard, Aleksandr Setkov, Angelique Loesch, and Romaric Audigier. Padim: a patch distribution modeling framework for anomaly detection and localization. In International Conference on Pattern Recognition, pages 475–489. Springer, 2021

    work page 2021

  7. [7]

    Automatic classification of defective photovoltaic module cells in electrolu- minescence images

    Sergiu Deitsch, Vincent Christlein, Stephan Berger, Claudia Buerhop-Lutz, Andreas Maier, Florian Gallwitz, and Christian Riess. Automatic classification of defective photovoltaic module cells in electrolu- minescence images. Solar Energy, 185:455–468, 2019

    work page 2019

  8. [8]

    Anomaly detection via reverse distillation from one-class embedding

    Hanqiu Deng and Xingyu Li. Anomaly detection via reverse distillation from one-class embedding. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9737–9746, 2022

    work page 2022

Show all 46 references
  1. [9]

    Memorizing normality to detect anomaly: Memory-augmented deep autoencoder for unsupervised anomaly detection

    Dong Gong, Lingqiao Liu, Vuong Le, Budhaditya Saha, Moussa Reda Mansour, Svetha Venkatesh, and Anton van den Hengel. Memorizing normality to detect anomaly: Memory-augmented deep autoencoder for unsupervised anomaly detection. InProceedings of the IEEE/CVF international confer...

    2019

  2. [10]

    Anomalygpt: Detecting industrial anomalies using large vision-language models

    Zhaopeng Gu, Bingke Zhu, Guibo Zhu, Yingying Chen, Ming Tang, and Jinqiao Wang. Anomalygpt: Detecting industrial anomalies using large vision-language models. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 1932–1940, 2024

    1932

  3. [11]

    Diad: A diffusion-based framework for multi-class anomaly detection

    Haoyang He, Jiangning Zhang, Hongxu Chen, Xuhai Chen, Zhishan Li, Xu Chen, Yabiao Wang, Chengjie Wang, and Lei Xie. Diad: A diffusion-based framework for multi-class anomaly detection. arXiv preprint arXiv:2312.06607, 2023

    2023 arXiv

  4. [12]

    Denoising diffusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. Advances in neural information processing systems, 33:6840–6851, 2020

    2020

  5. [13]

    Automated segmentation of macular edema in oct using deep neural networks

    Junjie Hu, Yuanyuan Chen, and Zhang Yi. Automated segmentation of macular edema in oct using deep neural networks. Medical image analysis, 55:216–227, 2019

    2019

  6. [14]

    Regis- tration based few-shot anomaly detection

    Chaoqin Huang, Haoyan Guan, Aofan Jiang, Ya Zhang, Michael Spratling, and Yan-Feng Wang. Regis- tration based few-shot anomaly detection. In European Conference on Computer Vision, pages 303–319. Springer, 2022

    2022

  7. [15]

    Adapting visual- language models for generalizable anomaly detection in medical images

    Chaoqin Huang, Aofan Jiang, Jinghao Feng, Ya Zhang, Xinchao Wang, and Yanfeng Wang. Adapting visual- language models for generalizable anomaly detection in medical images. arXiv preprint arXiv:2403.12570, 2024

    2024 arXiv

  8. [16]

    Winclip: Zero-/few-shot anomaly classification and segmentation

    Jongheon Jeong, Yang Zou, Taewan Kim, Dongqing Zhang, Avinash Ravichandran, and Onkar Dabeer. Winclip: Zero-/few-shot anomaly classification and segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 19606–19616, 2023

    2023

  9. [17]

    Identifying medical diagnoses and treatable diseases by image-based deep learning

    Daniel S Kermany, Michael Goldbaum, Wenjia Cai, Carolina CS Valentim, Huiying Liang, Sally L Baxter, Alex McKeown, Ge Yang, Xiaokang Wu, Fangbing Yan, et al. Identifying medical diagnoses and treatable diseases by image-based deep learning. cell, 172(5):1122–1131, 2018

    2018

  10. [18]

    Cifar-10 (canadian institute for advanced research), 2010

    Alex Krizhevsky, Vinod Nair, and Geoffrey Hinton. Cifar-10 (canadian institute for advanced research), 2010

    2010

  11. [19]

    Gradient-based learning applied to document recognition

    Yann LeCun, Léon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324, 1998. 11

    1998

  12. [20]

    Cutpaste: Self-supervised learning for anomaly detection and localization

    Chun-Liang Li, Kihyuk Sohn, Jinsung Yoon, and Tomas Pfister. Cutpaste: Self-supervised learning for anomaly detection and localization. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 9664–9674, 2021

    2021

  13. [21]

    Self- supervised anomaly detection, staging and segmentation for retinal images

    Yiyue Li, Qicheng Lao, Qingbo Kang, Zekun Jiang, Shiyi Du, Shaoting Zhang, and Kang Li. Self- supervised anomaly detection, staging and segmentation for retinal images. Medical Image Analysis, 87:102805, 2023

    2023

  14. [22]

    Classifier two sample test for video anomaly detections

    Yusha Liu, Chun-Liang Li, and Barnabás Póczos. Classifier two sample test for video anomaly detections. In BMVC, page 71, 2018

    2018

  15. [23]

    Cones: Concept neurons in diffusion models for customized generation

    Zhiheng Liu, Ruili Feng, Kai Zhu, Yifei Zhang, Kecheng Zheng, Yu Liu, Deli Zhao, Jingren Zhou, and Yang Cao. Cones: Concept neurons in diffusion models for customized generation. In International Conference on Machine Learning, pages 21548–21566. PMLR, 2023

    2023

  16. [24]

    On diffusion modeling for anomaly detection

    Victor Livernoche, Vineet Jain, Yashar Hezaveh, and Siamak Ravanbakhsh. On diffusion modeling for anomaly detection. arXiv preprint arXiv:2305.18593, 2023

    2023 arXiv

  17. [25]

    Boomerang: Local sampling on image manifolds using diffusion models

    Lorenzo Luzi, Ali Siahkoohi, Paul M Mayer, Josue Casco-Rodriguez, and Richard Baraniuk. Boomerang: Local sampling on image manifolds using diffusion models. arXiv preprint arXiv:2210.12100, 2022

    2022 arXiv

  18. [26]

    Sdedit: Guided image synthesis and editing with stochastic differential equations

    Chenlin Meng, Yutong He, Yang Song, Jiaming Song, Jiajun Wu, Jun-Yan Zhu, and Stefano Ermon. Sdedit: Guided image synthesis and editing with stochastic differential equations. In International Conference on Learning Representations, 2021

    2021

  19. [27]

    Deep learning for anomaly detection: A review

    Guansong Pang, Chunhua Shen, Longbing Cao, and Anton Van Den Hengel. Deep learning for anomaly detection: A review. ACM computing surveys (CSUR), 54(2):1–38, 2021

    2021

  20. [28]

    Learning transferable visual models from natural language supervision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learning transferable visual models from natural language supervision. In International conference on machine learning, pa...

    2021

  21. [29]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022

    2022

  22. [30]

    Towards total recall in industrial anomaly detection

    Karsten Roth, Latha Pemula, Joaquin Zepeda, Bernhard Schölkopf, Thomas Brox, and Peter Gehler. Towards total recall in industrial anomaly detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14318–14328, 2022

    2022

  23. [31]

    Dream- booth: Fine tuning text-to-image diffusion models for subject-driven generation

    Nataniel Ruiz, Yuanzhen Li, Varun Jampani, Yael Pritch, Michael Rubinstein, and Kfir Aberman. Dream- booth: Fine tuning text-to-image diffusion models for subject-driven generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 225...

    2023

  24. [32]

    A public fabric database for defect detection methods and results

    Javier Silvestre-Blanes, Teresa Albero-Albero, Ignacio Miralles, Rubén Pérez-Llorens, and Jorge Moreno. A public fabric database for defect detection methods and results. Autex Research Journal, 19(4):363–374, 2019

    2019

  25. [33]

    Real-world anomaly detection in surveillance videos

    Waqas Sultani, Chen Chen, and Mubarak Shah. Real-world anomaly detection in surveillance videos. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 6479–6488, 2018

    2018

  26. [34]

    Segmentation-based deep-learning approach for surface-defect detection

    Domen Tabernik, Samo Šela, Jure Skvarˇc, and Danijel Skoˇcaj. Segmentation-based deep-learning approach for surface-defect detection. Journal of Intelligent Manufacturing, 31(3):759–776, 2020

    2020

  27. [35]

    Revisiting reverse distillation for anomaly detection

    Tran Dinh Tien, Anh Tuan Nguyen, Nguyen Hoang Tran, Ta Duc Huy, Soan Duong, Chanh D Tr Nguyen, and Steven QH Truong. Revisiting reverse distillation for anomaly detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 24511–24520, 2023

    2023

  28. [36]

    Few-shot fast-adaptive anomaly detection

    Ze Wang, Yipin Zhou, Rui Wang, Tsung-Yu Lin, Ashish Shah, and Ser Nam Lim. Few-shot fast-adaptive anomaly detection. Advances in Neural Information Processing Systems, 35:4957–4970, 2022

    2022

  29. [37]

    Exploiting structural consistency of chest anatomy for unsupervised anomaly detection in radiography images

    Tiange Xiang, Yixiao Zhang, Yongyi Lu, Alan Yuille, Chaoyi Zhang, Weidong Cai, and Zongwei Zhou. Exploiting structural consistency of chest anatomy for unsupervised anomaly detection in radiography images. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024

    2024

  30. [38]

    Pushing the limits of fewshot anomaly detection in industry vision: Graphcore

    Guoyang Xie, Jinbao Wang, Jiaqi Liu, Feng Zheng, and Yaochu Jin. Pushing the limits of fewshot anomaly detection in industry vision: Graphcore. arXiv preprint arXiv:2301.12082, 2023

    2023 arXiv

  31. [39]

    Draem-a discriminatively trained reconstruction embedding for surface anomaly detection

    Vitjan Zavrtanik, Matej Kristan, and Danijel Sko ˇcaj. Draem-a discriminatively trained reconstruction embedding for surface anomaly detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 8330–8339, 2021

    2021

  32. [40]

    Reconstruction by inpainting for visual anomaly detection

    Vitjan Zavrtanik, Matej Kristan, and Danijel Sko ˇcaj. Reconstruction by inpainting for visual anomaly detection. Pattern Recognition, 112:107706, 2021

    2021

  33. [41]

    Expanding small-scale datasets with guided imagination

    Yifan Zhang, Daquan Zhou, Bryan Hooi, Kai Wang, and Jiashi Feng. Expanding small-scale datasets with guided imagination. Advances in Neural Information Processing Systems, 36, 2024. 12

    2024

  34. [42]

    Conditional prompt learning for vision- language models

    Kaiyang Zhou, Jingkang Yang, Chen Change Loy, and Ziwei Liu. Conditional prompt learning for vision- language models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 16816–16825, 2022

    2022

  35. [43]

    Encoding structure-texture relation with p-net for anomaly detection in retinal images

    Kang Zhou, Yuting Xiao, Jianlong Yang, Jun Cheng, Wen Liu, Weixin Luo, Zaiwang Gu, Jiang Liu, and Shenghua Gao. Encoding structure-texture relation with p-net for anomaly detection in retinal images. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 2...

    2020

  36. [44]

    Anomalyclip: Object-agnostic prompt learning for zero-shot anomaly detection

    Qihang Zhou, Guansong Pang, Yu Tian, Shibo He, and Jiming Chen. Anomalyclip: Object-agnostic prompt learning for zero-shot anomaly detection. arXiv preprint arXiv:2310.18961, 2023

    2023

  37. [45]

    Toward generalist anomaly detection via in-context residual learning with few-shot sample prompts

    Jiawen Zhu and Guansong Pang. Toward generalist anomaly detection via in-context residual learning with few-shot sample prompts. arXiv preprint arXiv:2403.06495, 2024

    2024 arXiv

  38. [46]

    [o] without flaw

    Yang Zou, Jongheon Jeong, Latha Pemula, Dongqing Zhang, and Onkar Dabeer. Spot-the-difference self-supervised pre-training for anomaly detection and segmentation. InEuropean Conference on Computer Vision, pages 392–408. Springer, 2022. 13 A Appendix / supplemental material A.1...

    2022

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.