Pith. sign in

REVIEW 3 major objections 6 minor 43 references

Unlocking the Potential of Reverse Distillation for Anomaly Detection

T0 review · 3 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read Adding an expert network and similarity-gated injection pushes reverse distillation to 99.0% pixel-level AUC on MVTec AD, with best localization among reverse-distillation methods on MPDD, BTAD, and VisA.

desk verdict A serious RD variant with consistent localization gains across four datasets, but the training objective is under-specified: the paper never states a detach on teacher features, which could undermine its central sensitivity mechanism, and the BTAD test-set alteration is not matched with re-run baselines. read the letter →

arxiv 2412.07579 v1 pith:GDGFDC76 submitted 2024-12-10 cs.CV

Xinyue Liu , Jianyuan Wang , Biao Leng , Shuo Zhang This is my paper · ORCID
classification cs.CV
keywords unsupervisedanomalydetectionknowledgedistillationreverselocalizationsynthetictrainingteacher-studentnetworksfeaturereconstructionindustrialdefect
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

Knowledge distillation is used for anomaly detection by training a student network to imitate a teacher on normal images and then flagging places where the imitation fails. The paper targets the Reverse Distillation variant, in which an encoder serves as teacher and a decoder as student, and argues that this design leaves two failure modes untreated: the teacher is not explicitly trained to react to anomalies, the student can over-generalize and reproduce anomalous features, and omitting skip connections loses fine detail and creates false alarms. To fix both, the paper proposes an Expert-Teacher-Student network in which a frozen expert encoder that sees only normal images supervises both the teacher and the student: the teacher is pushed to increase its cosine distance to the expert inside synthetic anomalies, and the student is pushed to emit normal features even from anomalous inputs. A Guided Information Injection module then uses high-level teacher–student similarity as attention to transfer lower-level details without leaking anomalies. The paper's core empirical claim is that this cuts missed detections and false positives and reaches 99.0% pixel-level AUC on MVTec AD, with the best localization among reverse-distillation methods on MPDD, BTAD, and VisA; the paper also notes its image-level detection on MPDD trails the MemKD baseline and that VisA masks were downsampled, which may affect very small anomalies.

What carries the argument

The load-bearing object is the Expert-Teacher-Student network, where the expert is a frozen reference of normality. The mechanism that carries the argument is the cosine-distance map $D^{n/a}_{TE,i}(h,w) = 1 - \cos\bigl(F^{n/a}_{T,i}(h,w), F^{n}_{E,i}(h,w)\bigr)$ between teacher and expert features; Eq. (2) trains that map to match the synthetic anomaly mask, so the teacher's sensitivity is optimized rather than assumed. The student is simultaneously denoised with the flattened-feature cosine losses in Eq. (5), and the second mechanism, Guided Information Injection, forms an attention map $\mathrm{Sim}^{i+1}$ from the cosine similarity of higher-level teacher and student features and uses it to blend fused teacher features with student features before the next decoder block: $F^{i+1}_{SSA} = \mathrm{Conv}_{3\times3}(F^{i+1}_{Tfuse}\odot \mathrm{Sim} + F^{i+1}_S \odot (1-\mathrm{Sim}))$. This similarity-gated soft skip connection is what lets low-level detail cross from teacher to student without carrying anomaly information.

What would settle it

A direct test: retrain the method with the texture-overlay synthesis replaced by a different anomaly generator, then compare per-category pixel-level AUC on the same test sets; a large drop on defect types that do not resemble the synthetic masks would show the teacher's sensitivity is bound to the synthetic distribution rather than general.

Watch

Extended reading notes

Core claim

The paper claims that the two prerequisites of Reverse Distillation—a teacher that separates normal from anomalous features and a student that always emits normal features—can be actively enforced rather than assumed. In its Expert-Teacher-Student network, the teacher $T$ and a frozen expert $E$ are WideResNet50 networks, the student $S$ is the symmetric upsampling counterpart, and $E$ sees only normal images. Training on pairs of normal and synthetically anomalous images, the teacher loss $L_{TE}$ (Eqs. 1–3) regresses the cosine-distance maps between teacher and expert features toward synthetic anomaly masks $M_{gt}$, making anomalous regions stand out while normal regions stay close to the expert. The student loss $L_S$ (Eqs. 4–6) makes $S$ reconstruct the normal features of both teacher and expert regardless of whether its input is normal or anomalous. Guided Information Injection, described in Algorithm 1, computes the cosine similarity between higher-level teacher and student features and uses it to gate lower-level teacher features fed into the student, replacing the rejected skip connection with a softer, anomaly-filtered transfer. The consequence, the paper reports, is that teacher–student disagreement concentrates in real anomalous regions: 99.0% pixel-level AUC on MVTec AD, the best localization among reverse-distillation methods on MPDD, BTAD, and VisA, and reduced false positives in normal regions.

Load-bearing premise

The load-bearing premise is that synthetic defects—random-shaped blobs generated with Perlin noise and overlaid with textures from an external dataset—teach the teacher to react to real industrial defects of any appearance; if a real defect looks too different from those synthetic ones, the trained sensitivity may not fire.

Editorial extensions

If this is right

  • Reverse-distillation training no longer has to assume the teacher catches anomalies: the expert-distillation loss explicitly shapes the teacher's cosine-distance map to synthetic anomaly masks, so the teacher is trained to be anomaly-sensitive.
  • The similarity-attention gate in Guided Information Injection provides a principled replacement for skip connections: low-level detail can be injected into the student while the attention map suppresses anomalous regions, reducing false positives in normal areas.
  • During inference the expert is removed and both teacher and student are frozen, so the method does not add storage or compute relative to the original Reverse Distillation at test time.
  • On MVTec AD the method reaches 99.0% pixel-level AUC, and among reverse-distillation methods it reports the best localization on MPDD, BTAD, and VisA; image-level detection is mixed, with the paper itself noting it falls short of MemKD on MPDD.
  • The ablations show both components matter: adding the Expert-Teacher-Student network alone mainly reduces missed detections, adding Guided Information Injection mainly suppresses background noise, and the full combination yields the best localization.

Reading between the lines

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

  • A direct stress test of the central transfer premise would train on texture-overlay synthetic anomalies and then evaluate on a held-out family of defects that is structurally different from those overlays (tiny scratches, low-contrast stains, glossy reflections), which the paper does not report.
  • Because Guided Information Injection computes its gate from high-level teacher–student similarity, anomalies visible mainly in low-level features may be under-injected; a multi-scale or low-level similarity gate is a natural variant the paper does not test.
  • The expert role is defined only by seeing normal images, so a lighter or domain-specific normal reference could replace the WideResNet50 expert, potentially lowering training cost while keeping the same dual-distillation signal.
  • A natural extension is to swap the texture-overlay synthesis for diffusion-generated anomalies; if the expert-distillation signal is what matters rather than the specific mask distribution, localization should survive in domains where texture-overlay priors are weak.
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 / 6 minor

Summary. The paper proposes RD-E, an anomaly-detection method built on the Reverse Distillation (RD) paradigm. It introduces a frozen expert encoder that is used during training to guide both a trainable teacher encoder and a trainable student decoder. Using synthetic anomalies generated from Perlin noise and DTD textures, the training losses (Eqs. 1-6) encourage the teacher to produce features that are distinct from the expert's features in anomalous regions, while encouraging the student to reconstruct normal-like features from both normal and anomalous teacher features. A Guided Information Injection (GII) module selectively transfers lower-level teacher features into the student, weighted by cosine similarity of higher-level features. At inference the expert is removed, so the method incurs no additional storage or compute cost. Experiments on MVTec AD, MPDD, BTAD, and VisA report improvements over RD and RD++ in pixel-level AUC and AP, with the strongest claimed gains on MVTec AD (P-AUC 99.0%) and on BTAD and MPDD. The paper includes ablations separating the contributions of the expert-based distillation and GII, and the code is publicly available.

Significance. If substantiated, the results would strengthen the RD paradigm and give a practical method with no additional inference overhead. The paper is transparent about its reliance on synthetic anomalies, and it evaluates on multiple public benchmarks rather than only synthetic data. The ablation study confirms that both the expert-based distillation and GII matter. The main reservations are that the training objective is underspecified with respect to gradient flow into the teacher, and the BTAD test set is altered without full disclosure; these issues affect the interpretation of the central claims.

major comments (3)
  1. [Method, Eqs. (1)-(6)] The training objective as written does not specify whether teacher features are detached before being fed into the student and before being used as targets in L_ST. Since the teacher is trainable and the student consumes teacher features as input, gradients from L_S can flow into the teacher. In anomalous regions, L_ST then provides an incentive for the teacher to output features that resemble the expert's normal features, which directly opposes the sensitivity loss L_TE that pushes the teacher to differ from the expert. The paper never states that a stop-gradient is applied, so the claim that RD-E 'enhances the teacher's differentiation' is not implied by the loss as written. Please clarify whether teacher features are detached, and if not, analyze the interaction of the two gradient contributions with either a gradient analysis or a small experiment.
  2. [Supplementary, Details of Datasets (BTAD)] The supplementary text states that 'some misclassified images' were removed from the original BTAD dataset before experiments, but it does not report how many images were removed, from which classes, or by what criterion. BTAD has only 736 test images across three classes, so even a small number of removed images can shift the reported metrics. Since the paper compares against published RD and RD++ baselines, the authors should report the exact number of removed images per class and also provide results on the original, unmodified BTAD test set so that the comparison with prior work is meaningful.
  3. [Experiments, Tables 1-4] All reported results are from a single training run, with no standard deviations or multiple seeds. The central claim of surpassing RD and RD++ is based on metric differences that are sometimes small (e.g., MVTec P-AUC 99.0 vs. 98.2 for RD++). Without an estimate of run-to-run variance, it is unclear whether these differences are significant. Please provide results over at least three seeds with mean and standard deviation for the main comparison tables, or a suitable statistical test.
minor comments (6)
  1. [Abstract] The phrase 'To addresses this problem' should be 'To address this problem'.
  2. [Section Revisiting Reverse Distillation] The sentence 'The student is promised to generate anomaly-free features' should read 'The student is assumed to generate anomaly-free features'.
  3. [Eq. (5)] The combined notation L^i_SE/ST is ambiguous; the two terms should be defined separately as L^i_SE and L^i_ST.
  4. [Main text, Anomaly Localization paragraph] The text refers to 'Tables 3 and 10' for MPDD and BTAD, but the BTAD result is presented in Table 4 of the main text; the table numbering should be made consistent.
  5. [Main text, Ablation Study on GII] The text refers to 'Table 12' for the GII ablation results, which is a supplementary table; it should cite the corresponding main-text table (Table 7).
  6. [Supplementary, Details of Anomaly Synthesis] The sentence 'the corresponding synthetic anomalous image Ia for a given normal image Ia is expressed as' contains a typo; it should read 'for a given normal image I_n'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the method is an empirical benchmark contribution, and its internal losses are training objectives evaluated against independent real-world anomaly masks.

full rationale

The paper makes no first-principles theoretical claim; its central assertions are empirical benchmark comparisons on MVTec AD, MPDD, BTAD, and VisA. The only internal regression target is the teacher-sensitivity loss in Eq. (2), which fits the teacher-expert cosine distance map to the synthetic anomaly mask. This is a supervised auxiliary training objective, not a prediction of real anomalies: the reported pixel-level AUC, AP, and PRO are computed on held-out test images using independently labeled real ground-truth masks, so the benchmark result is not forced by the training loss. Likewise, the student denoising loss L_S directly optimizes cosine similarity to teacher/expert normal features, but the final evaluation uses the frozen teacher and student at inference on separate test data. The one self-citation (Liu et al. 2024) appears only in the related-work discussion of asymmetric teacher-student networks and is not load-bearing for any architectural choice or performance claim. External components such as RD, DRÆM synthetic anomalies, DTD textures, and the ReContrast flatten operation are explicitly and properly cited. No uniqueness theorem, no fitted parameter renamed as prediction, and no derivation that reduces by definition to its inputs was found. The paper is therefore self-contained against external benchmarks for the claims it makes.

Assumptions & free parameters 4 free parameters · 5 assumptions · 1 invented entities

The central claim rests on a pre-trained backbone, on the validity of synthetic anomalies as a training proxy, on the cosine-distance objective as an anomaly score, and on the attention-gating assumption in GII. The only invented component is the frozen expert network, whose benefit is evidenced only by in-paper ablations. There are several hand-chosen hyperparameters (sigma, opacity range, foreground threshold, three layers) that affect the reported numbers.

free parameters (4)
  • Gaussian smoothing sigma = 4
    Used to smooth anomaly maps at inference; chosen by hand, affects localization scores.
  • Synthetic anomaly opacity beta = sampled in [0.15, 1]
    Follows DeSTSeg; controls blending of texture into normal image during synthetic anomaly generation.
  • Foreground mask threshold = not specified
    Used to restrict synthetic anomalies to object foreground in object categories; threshold applied to binarized images, a crude heuristic.
  • Number of distilled feature layers = 3
    Only first three stages of WideResNet50 are used for anomaly maps; choice affects detail and computational cost.
assumptions (5)
  • domain assumption A WideResNet50 pre-trained on ImageNet provides a suitable feature space for industrial anomaly detection.
    The teacher, student, and expert all start from ImageNet weights; the method assumes these features transfer to industrial defects.
  • ad hoc to paper Synthetic anomalies generated by Perlin noise and DTD textures are a valid proxy for real anomalies during training.
    The paper's training uses these synthetic anomalies and their masks; no evidence that the learned sensitivity transfers to real anomaly types not resembling the synthetic ones.
  • domain assumption The cosine distance between teacher and student features is a reliable anomaly score that can be trained directly (Eq. 2 uses the mask as regression target).
    The teacher loss explicitly regresses this distance to the synthetic mask; the method assumes this training objective optimizes the final detection metric.
  • ad hoc to paper Higher-level feature similarity between teacher and student reliably localizes anomalies and can gate low-level feature injection.
    GII uses this similarity as attention; the paper provides ablations but no principled derivation of why this gate prevents anomaly leakage.
  • standard math Standard backpropagation and cosine similarity definitions.
    Background used in all losses.
invented entities (1)
  • Expert network (frozen copy of the initial teacher)
    purpose: Provides a frozen 'normal' feature reference to train the teacher's anomaly sensitivity and the student's denoising; removed at inference.
    The expert is an internal architectural addition with no falsifiable external prediction; its benefit is supported only by in-paper ablations.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Unlocking the Potential of Reverse Distillation for Anomaly Detection." pith.science (2026). https://pith.science/paper/GDGFDC76

@misc{pith2026241207579,
  author       = {Pith},
  title        = {Pith review of: Unlocking the Potential of Reverse Distillation for Anomaly Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GDGFDC76}},
  note         = {Machine review of arXiv:2412.07579}
}
read the original abstract

Knowledge Distillation (KD) is a promising approach for unsupervised Anomaly Detection (AD). However, the student network's over-generalization often diminishes the crucial representation differences between teacher and student in anomalous regions, leading to detection failures. To addresses this problem, the widely accepted Reverse Distillation (RD) paradigm designs the asymmetry teacher and student, using an encoder as teacher and a decoder as student. Yet, the design of RD does not ensure that the teacher encoder effectively distinguishes between normal and abnormal features or that the student decoder generates anomaly-free features. Additionally, the absence of skip connections results in a loss of fine details during feature reconstruction. To address these issues, we propose RD with Expert, which introduces a novel Expert-Teacher-Student network for simultaneous distillation of both the teacher encoder and student decoder. The added expert network enhances the student's ability to generate normal features and optimizes the teacher's differentiation between normal and abnormal features, reducing missed detections. Additionally, Guided Information Injection is designed to filter and transfer features from teacher to student, improving detail reconstruction and minimizing false positives. Experiments on several benchmarks prove that our method outperforms existing unsupervised AD methods under RD paradigm, fully unlocking RD's potential.

Figures

Figures reproduced from arXiv: 2412.07579 by the authors.

Figure 1
Figure 1. Anomaly localization examples. Our method re [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Schematic diagram of the framework and data flow [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Overview of our proposed method. (a) shows the overall architecture and training process of our designed Expert [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (12 more)
Figure 4
Figure 4. Figure 4: (a) Cosine distance maps between features of [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: Inference procedure of our proposed method. The [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Visualization of ablation study on network composition. From top to bottom: the input image, the ground truth masks, [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 8
Figure 8. Figure 8: Experimental Setup Details of Datasets MVTec AD MVTec AD (Bergmann et al. 2019) is a com￾monly used benchmark for unsupervised AD. It consists of 15 categories of industrial images, including 5 texture cat￾egories and 10 object categories. The training set contains 362…
Figure 7
Figure 7. Figure 7: Anomaly synthesis of texture images in MVTec AD. [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 8
Figure 8. Figure 8: Anomaly synthesis of object images in MVTec AD. [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]
Figure 9
Figure 9. Figure 9: Visualization of ablation study on network composition on MPDD and BTAD. From top to bottom: the input image, [PITH_FULL_IMAGE:figures/full_fig_p013_9.png]
Figure 10
Figure 10. Figure 10: Missed detections of RD [PITH_FULL_IMAGE:figures/full_fig_p015_10.png]
Figure 11
Figure 11. Figure 11: False positives of RD. Input GT RD RD++ Ours [PITH_FULL_IMAGE:figures/full_fig_p016_11.png]
Figure 12
Figure 12. Figure 12: Visualization of anomaly localization on MPDD. [PITH_FULL_IMAGE:figures/full_fig_p016_12.png]
Figure 13
Figure 13. Figure 13: Visualization of anomaly localization on BTAD. [PITH_FULL_IMAGE:figures/full_fig_p017_13.png]
Figure 14
Figure 14. Figure 14: Visualization of anomaly localization on VisA. [PITH_FULL_IMAGE:figures/full_fig_p018_14.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 26 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Akcay, S.; Atapour-Abarghouei, A.; and Breckon, T. P. 2019. Ganomaly: Semi-supervised anomaly detection via adversarial training. In Computer Vision--ACCV 2018: 14th Asian Conference on Computer Vision, Perth, Australia, December 2--6, 2018, Revised Selected Papers, Part III 14, 622--637. Springer

    work page 2019

  4. [4]

    Bae, J.; Lee, J.-H.; and Kim, S. 2023. Pni: industrial anomaly detection using position and neighborhood information. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 6373--6383

    work page 2023

  5. [5]

    Bergmann, P.; Fauser, M.; Sattlegger, D.; and Steger, C. 2019. MVTec AD--A comprehensive real-world dataset for unsupervised anomaly detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 9592--9600

    2019

  6. [6]

    Bergmann, P.; Fauser, M.; Sattlegger, D.; and Steger, C. 2020. Uninformed students: Student-teacher anomaly detection with discriminative latent embeddings. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 4183--4192

    work page 2020

  7. [7]

    Bergmann, P.; L \"o we, S.; Fauser, M.; Sattlegger, D.; and Steger, C. 2018. Improving unsupervised defect segmentation by applying structural similarity to autoencoders. arXiv preprint arXiv:1807.02011

    arXiv 2018

  8. [8]

    Cimpoi, M.; Maji, S.; Kokkinos, I.; Mohamed, S.; and Vedaldi, A. 2014. Describing textures in the wild. In Proceedings of the IEEE conference on computer vision and pattern recognition, 3606--3613

    2014

Show all 43 references
  1. [9]

    Defard, T.; Setkov, A.; Loesch, A.; and Audigier, R. 2021. Padim: a patch distribution modeling framework for anomaly detection and localization. In International Conference on Pattern Recognition, 475--489. Springer

    2021

  2. [10]

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

    2022

  3. [11]

    Deng, J.; Dong, W.; Socher, R.; Li, L.-J.; Li, K.; and Fei-Fei, L. 2009. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, 248--255. Ieee

    2009

  4. [12]

    Gu, Z.; Liu, L.; Chen, X.; Yi, R.; Zhang, J.; Wang, Y.; Wang, C.; Shu, A.; Jiang, G.; and Ma, L. 2023. Remembering Normality: Memory-guided Knowledge Distillation for Unsupervised Anomaly Detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 16...

    2023

  5. [13]

    Gudovskiy, D.; Ishizaka, S.; and Kozuka, K. 2022. Cflow-ad: Real-time unsupervised anomaly detection with localization via conditional normalizing flows. In Proceedings of the IEEE/CVF winter conference on applications of computer vision, 98--107

    2022

  6. [14]

    Guo, H.; Ren, L.; Fu, J.; Wang, Y.; Zhang, Z.; Lan, C.; Wang, H.; and Hou, X. 2023. Template-guided Hierarchical Feature Restoration for Anomaly Detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 6447--6458

    2023

  7. [15]

    Guo, J.; Jia, L.; Zhang, W.; Li, H.; et al. 2024. Recontrast: Domain-specific anomaly detection via contrastive reconstruction. Advances in Neural Information Processing Systems, 36

    2024

  8. [16]

    H.; Bae, K.; and Kang, B

    Hyun, J.; Kim, S.; Jeon, G.; Kim, S. H.; Bae, K.; and Kang, B. J. 2024. ReConPatch: Contrastive patch representation learning for industrial anomaly detection. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2052--2061

    2024

  9. [17]

    Jezek, S.; Jonak, M.; Burget, R.; Dvorak, P.; and Skotak, M. 2021. Deep learning-based defect detection of metal parts: evaluating current methods in complex conditions. In 2021 13th International congress on ultra modern telecommunications and control systems and workshops (I...

    2021

  10. [18]

    Jiang, Y.; Cao, Y.; and Shen, W. 2023. A masked reverse knowledge distillation method incorporating global and local information for image anomaly detection. Knowledge-Based Systems, 280: 110982

    2023

  11. [19]

    Li, C.-L.; Sohn, K.; Yoon, J.; and Pfister, T. 2021. Cutpaste: Self-supervised learning for anomaly detection and localization. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 9664--9674

    2021

  12. [20]

    Li, H.; Chen, Z.; Xu, Y.; and Hu, J. 2024. Hyperbolic Anomaly Detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 17511--17520

    2024

  13. [21]

    Li, H.; Hu, J.; Li, B.; Chen, H.; Zheng, Y.; and Shen, C. 2023. Target before shooting: Accurate anomaly detection and localization under one millisecond via cascade patch retrieval. arXiv preprint arXiv:2308.06748

    2023 arXiv

  14. [22]

    Lin, J.; and Yan, Y. 2024. A Comprehensive Augmentation Framework for Anomaly Detection. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, 8742--8749

    2024

  15. [23]

    Liu, T.; Li, B.; Du, X.; Jiang, B.; Geng, L.; Wang, F.; and Zhao, Z. 2023 a . Fair: frequency-aware image restoration for industrial visual anomaly detection. arXiv preprint arXiv:2309.07068

    2023 arXiv

  16. [24]

    Liu, X.; Wang, J.; Leng, B.; and Zhang, S. 2024. Dual-modeling decouple distillation for unsupervised anomaly detection. In Proceedings of the 32nd ACM International Conference on Multimedia, 5035--5044

    2024

  17. [25]

    Liu, Z.; Zhou, Y.; Xu, Y.; and Wang, Z. 2023 b . Simplenet: A simple network for image anomaly detection and localization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 20402--20411

    2023

  18. [26]

    Mishra, P.; Verk, R.; Fornasier, D.; Piciarelli, C.; and Foresti, G. L. 2021. VT-ADL: A vision transformer network for image anomaly detection and localization. In 2021 IEEE 30th International Symposium on Industrial Electronics (ISIE), 01--06. IEEE

    2021

  19. [27]

    Perlin, K. 1985. An image synthesizer. In Proceedings of the 12th Annual Conference on Computer Graphics and Interactive Techniques, SIGGRAPH '85, 287–296. ISBN 0897911660

    1985

  20. [28]

    Roth, K.; Pemula, L.; Zepeda, J.; Sch \"o lkopf, B.; Brox, T.; and Gehler, P. 2022. Towards total recall in industrial anomaly detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 14318--14328

    2022

  21. [29]

    Rudolph, M.; Wehrbein, T.; Rosenhahn, B.; and Wandt, B. 2023. Asymmetric student-teacher networks for industrial anomaly detection. In Proceedings of the IEEE/CVF winter conference on applications of computer vision, 2592--2602

    2023

  22. [30]

    H.; and Rabiee, H

    Salehi, M.; Sadjadi, N.; Baselizadeh, S.; Rohban, M. H.; and Rabiee, H. R. 2021. Multiresolution knowledge distillation for anomaly detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 14902--14912

    2021

  23. [31]

    Tang, T.-W.; Kuo, W.-H.; Lan, J.-H.; Ding, C.-F.; Hsu, H.; and Young, H.-T. 2020. Anomaly detection neural network with dual auto-encoders GAN and its industrial inspection applications. Sensors, 20(12): 3336

    2020

  24. [32]

    D.; Nguyen, A

    Tien, T. D.; Nguyen, A. T.; Tran, N. H.; Huy, T. D.; Duong, S.; Nguyen, C. D. T.; and Truong, S. Q. 2023. Revisiting reverse distillation for anomaly detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 24511--24520

    2023

  25. [33]

    Wang, G.; Han, S.; Ding, E.; and Huang, D. 2021. Student-Teacher Feature Pyramid Matching for Anomaly Detection. In 32nd British Machine Vision Conference 2021, BMVC 2021, Online, November 22-25, 2021 , 306. BMVA Press

    2021

  26. [34]

    Zagoruyko, S.; and Komodakis, N. 2016. Wide residual networks. arXiv preprint arXiv:1605.07146

    2016 arXiv

  27. [35]

    Zavrtanik, V.; Kristan, M.; and Sko c aj, D. 2021. Draem-a discriminatively trained reconstruction embedding for surface anomaly detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 8330--8339

    2021

  28. [36]

    Zhang, J.; Suganuma, M.; and Okatani, T. 2024. Contextual affinity distillation for image anomaly detection. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 149--158

    2024

  29. [37]

    Zhang, X.; Li, N.; Li, J.; Dai, T.; Jiang, Y.; and Xia, S.-T. 2023 a . Unsupervised surface anomaly detection with diffusion probabilistic model. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 6782--6791

    2023

  30. [38]

    Zhang, X.; Li, S.; Li, X.; Huang, P.; Shan, J.; and Chen, T. 2023 b . Destseg: Segmentation guided denoising student-teacher for anomaly detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 3914--3923

    2023

  31. [39]

    Zhang, X.; Xu, M.; and Zhou, X. 2024. RealNet: A feature selection network with realistic synthetic anomaly for anomaly detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 16699--16708

    2024

  32. [40]

    Zhou, Q.; He, S.; Liu, H.; Chen, T.; and Chen, J. 2022. Pull & push: Leveraging differential knowledge distillation for efficient unsupervised anomaly detection and localization. IEEE Transactions on Circuits and Systems for Video Technology

    2022

  33. [41]

    Zhou, Y.; Xu, X.; Song, J.; Shen, F.; and Shen, H. T. 2024. MSFlow: Multiscale Flow-Based Framework for Unsupervised Anomaly Detection. IEEE Transactions on Neural Networks and Learning Systems

    2024

  34. [42]

    Zhang, X.; Li, S.; Li, X.; Huang, P.; Shan, J.; and Chen, T. 2023. Destseg: Segmentation guided denoising student-teacher for anomaly detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 3914--3923

    2023

  35. [43]

    Zou, Y.; Jeong, J.; Pemula, L.; Zhang, D.; and Dabeer, O. 2022. Spot-the-difference self-supervised pre-training for anomaly detection and segmentation. In European Conference on Computer Vision, 392--408. Springer

    2022

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.