Pith. sign in

REVIEW 4 major objections 6 minor 38 references

Tuned Reverse Distillation: Enhancing Multimodal Industrial Anomaly Detection with Crossmodal Tuners

T0 review · 4 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 A multi-branch reverse distillation with crossmodal filters and amplifiers solves the anomaly-smoothing problem in multimodal knowledge-distillation anomaly detection, reporting the best I-AUC (96.2%) and PRO (97.7%) on MVTec 3D-AD.

desk verdict A solid multi-branch distillation architecture for multimodal AD whose real contribution is the Crossmodal Filter/Amplifier design; the SOTA headline is fragile because the margins are sub-point and the hyperparameters were tuned on the test set. read the letter →

arxiv 2412.08949 v4 pith:MCD4D4QC submitted 2024-12-12 cs.CV

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

The paper claims that the common practice of fusing RGB and 3D information before or during knowledge distillation smoothes away anomalies that appear in only one modality, producing false negatives. To fix this, it proposes Tuned Reverse Distillation (TRD), which gives each modality its own reverse-distillation branch and adds two Crossmodal Tuners that let branches exchange information without diluting anomaly signals. On MVTec 3D-AD the method reports the highest image-level AUC (96.2%) and PRO (97.7%) among compared methods, and the best average performance (96.6%) across MVTec 3D-AD and Eyecandies among feature-learning methods. If correct, this establishes multi-branch distillation with crossmodal filtering and amplification as the leading approach for unsupervised multimodal industrial anomaly detection and localization.

What carries the argument

The load-bearing object is the pair of Crossmodal Tuners inserted into each reverse-distillation branch. The Crossmodal Filter uses a Bottleneck Projection—convolutional downsampling to 8x8 followed by deconvolutional restoration, trained to align with the other modality's teacher features—to strip anomaly content from the other modality before it is concatenated with the branch's own features and passed through the OCBE compressor. The Crossmodal Amplifier uses an Inverted Bottleneck Projection, which widens channels before compressing them, to map the other modality's features into the branch's feature space; because it is trained only on normal features, anomalies in the other modality survive as discrepancies that are fused into the student output via learned softmax weights. Together they realize crossmodal mapping inside a multi-branch reverse distillation, letting each branch see both its own and the other modality's anomalies.

What would settle it

Take a paired sample where an anomaly is clearly visible in the depth map but invisible in RGB, feed it through the RGB branch with the Crossmodal Amplifier active, and check whether the RGB branch's cosine-distance anomaly map fires at the anomaly location; if it stays flat while the depth branch detects it, the crossmodal mechanism is not actually transferring anomaly information and the reported gains would come from multi-branch fusion alone.

Watch

Extended reading notes

Core claim

The central claim is that the anomaly-smoothing problem in KD-based multimodal anomaly detection is caused by fusing modalities too early, and that a multi-branch design in which each modality is distilled independently, guided by two crossmodal mechanisms, resolves it. The Crossmodal Filter compresses the other modality's features through an 8x8 bottleneck so that only normal content passes into the student decoder, while the Crossmodal Amplifier expands then compresses the other modality's features so that anomalies in that modality create a detectable mismatch in this branch. Each branch thus reconstructs normality for its own modality and simultaneously becomes sensitive to anomalies in the other modality, and the per-branch anomaly maps are normalized and summed at inference. The paper reports that this design outperforms prior fusion-based KD methods and crossmodal feature mapping on MVTec 3D-AD and Eyecandies, with an average I-AUC of 96.2% and a PRO of 97.7% on MVTec 3D-AD.

Load-bearing premise

The load-bearing premise is that compressing one modality's features down to 8x8 and restoring them removes defects while preserving normality, and that depth and normal maps behave like ordinary photos under this treatment.

Editorial extensions

If this is right

  • Detection of single-modality anomalies improves: defects visible only in depth or only in RGB are no longer washed out by the normal modality during fusion.
  • Anomaly localization becomes more reliable on datasets where modalities disagree, since each branch can flag the other branch's anomalies before final map summation.
  • The method is efficient enough for near-real-time inspection: the paper reports 21.7 fps under the same hardware setup, higher than the compared feature-learning baselines.
  • The symmetric design means it can extend to any aligned modality pair or group, not just RGB and depth.

Reading between the lines

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

  • The same filter/amplifier mechanism could transfer to other modality pairs such as RGB and thermal imaging or RGB and surface-normal maps, though the 8x8 bottleneck size may need per-pair tuning because feature-space alignment differs across modalities.
  • The crossmodal filter's suppression of anomalies could be tested explicitly by injecting synthetic anomalies into one modality at inference and measuring whether the other branch's student features remain clean; the paper's ablations only test this indirectly through end metrics.
  • Because the tuners are lightweight CNN layers, the architecture might carry over to video anomaly detection by treating optical flow or audio as the second modality, with temporal alignment as an added requirement.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes Tuned Reverse Distillation (TRD), a knowledge-distillation method for unsupervised multimodal industrial anomaly detection. TRD assigns an independent reverse-distillation branch to each modality (RGB and depth), and introduces two Crossmodal Tuners: a Crossmodal Filter, which compresses features from the other modality through a bottleneck before feeding them to the student decoder, and a Crossmodal Amplifier, which maps features from the other modality into the current branch to amplify anomaly signals. The training losses combine decoder cosine-similarity losses, filter alignment losses, and amplifier mapping/consistency losses, all trained on normal samples only. Experiments on MVTec 3D-AD and Eyecandies report average I-AUC, P-AUC, PRO, and AP numbers, with the paper claiming state-of-the-art performance in both detection and localization, plus a higher frame rate than comparable feature-learning baselines. Code is made available.

Significance. If the empirical claims hold, TRD is a meaningful contribution to multimodal anomaly detection: the multi-branch design directly targets the anomaly-smoothing failure of early fusion, the two Crossmodal Tuners are clearly motivated, and the ablations in Tables V-VIII show that each component contributes to the reported performance. The paper is also reproducible in spirit: code is released, the training objective uses only normal samples, and the architecture is described in enough detail to re-implement. The main significance, however, rests on the claim that TRD 'consistently outperforms' prior methods; that claim currently depends on very narrow margins and on hyperparameters selected on the test set, so the scientific value will be determined by whether the authors can supply statistical evidence and a clean model-selection protocol.

major comments (4)
  1. [§V-B, Tables I-III] The central SOTA claim rests on average margins of 0.1-0.3 percentage points: I-AUC 96.2 vs 96.0 for CFM, PRO 97.7 vs 97.6 for Shape-guided/MMRD, and Table III average 96.6 vs 96.3 for MMRD. No error bars, multiple seeds, or significance tests are reported. These margins are within typical run-to-run variability for deep anomaly-detection models, so the claim that TRD 'consistently outperforms all other methods' is not yet supported. Please report mean and standard deviation over at least 3-5 random seeds and, for the key comparisons, a paired significance test or confidence interval.
  2. [§V-C, Tables VI-VII and §V-A] The Bottleneck Projection downsampling size (8×8) and Inverted Bottleneck Projection expansion factor (2×) are selected by comparing I-AUC and PRO on the MVTec 3D-AD test set, and no held-out validation split is described. The Gaussian smoothing sigma (σ=4) is likewise fixed without a validation-based justification. Selecting hyperparameters on the test set makes the reported test metrics optimistic and not unbiased estimates of generalization. Please use a separate validation split for model selection and report the test performance of the configuration selected on validation only.
  3. [Table III caption and MMRD*] Table III claims that all methods are compared 'under the same hardware setup', yet MMRD is marked as having no reproducible code. If the MMRD numbers are taken from the original publication rather than rerun by the authors, the same-hardware claim is inaccurate and the comparison is not controlled. Please clarify the source of every baseline number, and either provide a rerun of MMRD or explicitly qualify the comparison as using published results.
  4. [Table I, Rope and Dowel rows] The word 'consistently' overstates the per-category results. On Rope, TRD attains I-AUC 83.3 versus 95.0 for CFM; on Dowel, 90.1 versus 92.4. The average improvement is carried by other categories, so the claim that TRD 'consistently outperforms all other methods' is not accurate at the category level. Please either temper the claim or provide an analysis of where the average gains come from.
minor comments (6)
  1. [§IV-B.2 and Eq. (3)] Eq. (3) defines F_A_CA as a weighted average of the decoder output and the inverted-bottleneck output, but the text and Fig. 5 caption say the amplified features are 'added to the student decoder output'. Please align the wording with the equation.
  2. [§IV-B.1 and Fig. 4] The Bottleneck Projection is described as downsampling encoder features to 8×8, but the three encoder feature levels F_B_E_i have different spatial resolutions. Please clarify whether each level is independently resized to 8×8 or whether only a particular level is used.
  3. [Table II, CFM row] The PRO entry for one category is reported as '88.0/96.5', which appears to be two values in a single cell. Please correct this formatting error.
  4. [Table IV] The column header 'P-PRO' appears to be a typo for 'PRO'.
  5. [§III and Tables VI-VII] There are several typographical errors: 'Bottlneck' in Tables VI and VII and the surrounding text, 'brefore' in Section III, and the anomaly-map definition 'M = P(1 − Sim(FT , FS))' uses an unexplained symbol P and a stray period. Please proofread these passages.
  6. [§V-A] Please specify how the validation set is used for Eyecandies as well as MVTec 3D-AD, since Eq. (11) normalizes anomaly maps using normal samples from the validation set and the paper does not state the Eyecandies validation protocol explicitly.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central claims are empirical and evaluated on held-out anomalous test samples, with no fitted parameter renamed as a prediction and no load-bearing self-citation chain.

full rationale

The paper is an empirical architecture paper, not a derivation. Its central claim that TRD outperforms prior methods on MVTec 3D-AD and Eyecandies is supported by test-set metrics computed from held-out anomalous samples after training only on normal samples. The training objectives in Eqs. (4)-(9) are cosine-similarity losses between teacher and student features, and the anomaly maps in Eq. (10) compare teacher features against student/crossmodal outputs at inference; the anomaly information used for evaluation is not part of any training loss. The Crossmodal Filter and Amplifier are trained only on normal data, and their effect is assessed by ablations on the test set, so the reported improvement is an empirical measurement rather than an identity forced by construction. The validation-set normalization in Eq. (11) uses only normal samples and is a standard calibration step, not a fitted prediction target. The paper contains no uniqueness theorem, no claim that a choice is forced by prior work, and no load-bearing self-citation: all cited baselines (AST, CFM, MMRD, RD, M3DM, etc.) are external prior works, and the authors do not cite their own prior results as the justification for the method. The main legitimate concern is that Bottleneck Projection size (Table VI) and channel expansion factor (Table VII) are selected by comparing test-set metrics, and the reported SOTA margins over the runner-up are small (e.g., I-AUC 96.2 vs 96.0; PRO 97.7 vs 97.6) with no error bars or significance tests. That is a statistical-reliability and benchmark-selection issue, not circularity: the paper's predictions do not reduce to its inputs by definition, and the mechanism would still be testable even if the margins are within run-to-run variance. Therefore the circularity score is 0.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central mechanism rests on two transferred intuitions from single-modality RD: bottleneck compression separates normal from abnormal feature content, and cosine feature distance is a valid anomaly score. The method also assumes ImageNet RGB features transfer to depth and normal maps. The main tuned choices are bottleneck sizes and loss weights selected by hand or by ablation. No new physical entities are introduced.

free parameters (4)
  • Bottleneck Projection downsampling size = 8x8
    Selected by ablation in Table VI on MVTec 3D-AD; 8x8 gives the best average score and is used in the final model.
  • Inverted Bottleneck Projection channel expansion = 2x
    Selected by ablation in Table VII; 2x is chosen as the best balance between localization performance and computational cost.
  • Loss term weights = 1.0 for each loss component
    All loss components in Eq. 9 are assigned weight 1.0 by hand, with no tuning reported.
  • Anomaly map Gaussian smoothing sigma = 4
    Used during inference as described in Section V-A; a hand-chosen smoothing parameter.
assumptions (5)
  • domain assumption ImageNet-pretrained WideResNet-50 features transfer to depth and normal maps.
    The same frozen teacher encoder is used for RGB and 3D inputs (Section IV-A, Fig. 3).
  • domain assumption OCBE-style bottleneck compression removes anomaly information while preserving normal information.
    Crossmodal Filter and Amplifier rely on this inherited RD behavior (Sections III and IV-B).
  • domain assumption Cosine distance between teacher and tuned student features marks anomalous regions.
    The anomaly map is defined as 1 minus cosine similarity in Eq. 10, inheriting the standard RD assumption.
  • domain assumption Validation-set normal samples give reliable per-modality normalization statistics.
    Inference normalization in Eq. 11 uses mean and standard deviation computed from validation normal samples (Section IV-D).
  • domain assumption Compared baseline numbers are comparable despite differing preprocessing and code availability.
    Table III reruns AST and CFM under the same hardware, but MMRD numbers are taken from its paper because its code is unavailable (Section V-B3).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Tuned Reverse Distillation: Enhancing Multimodal Industrial Anomaly Detection with Crossmodal Tuners." pith.science (2026). https://pith.science/paper/MCD4D4QC

@misc{pith2026241208949,
  author       = {Pith},
  title        = {Pith review of: Tuned Reverse Distillation: Enhancing Multimodal Industrial Anomaly Detection with Crossmodal Tuners},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MCD4D4QC}},
  note         = {Machine review of arXiv:2412.08949}
}
read the original abstract

Knowledge distillation (KD) has been widely studied in unsupervised image Anomaly Detection (AD), but its application to unsupervised multimodal AD remains underexplored. Existing KD-based methods for multimodal AD that use fused multimodal features to obtain teacher representations face challenges. Anomalies that only exist in one modality may not be effectively captured in the fused teacher features, leading to detection failures. Besides, these methods do not fully leverage the rich intra- and inter-modality information that are critical for effective anomaly detection. In this paper, we propose Tuned Reverse Distillation (TRD) based on Multi-branch design to realize Multimodal Industrial AD. By assigning independent branches to each modality, our method enables finer detection of anomalies within each modality. Furthermore, we enhance the interaction between modalities during the distillation process by designing two Crossmodal Tuners including Crossmodal Filter and Amplifier. With the idea of crossmodal mapping, the student network is allowed to better learn normal features while anomalies in all modalities are ensured to be effectively detected. Experimental verifications on multimodal AD datasets demonstrate that our method achieves state-of-the-art performance in multimodal anomaly detection and localization. Code is available at https://github.com/hito2448/TRD.

Figures

Figures reproduced from arXiv: 2412.08949 by the authors.

Figure 1
Figure 1. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Comparison of our proposed TRD with previous KD-based methods. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Information flow of the branch for Modality A during training. This branch consists of a frozen teacher encoder, a trainable student decoder, and two Crossmodal Tuners including Crossmodal Filter and Crossmodal Amplifier. Crossmodal Filter incorporates mapped features from Modality B to guide the student decoder of Modality A in generating anomaly-free features. Crossmodal Amplifier injects anomaly signals from Moda… view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: Illustration of the Crossmodal Filter within the branch of Modality A, which consists of two components including Bottleneck Projection and OCBE. First, the encoder features from the other modality (Modality B) F B E i , i = {1, 2, 3} are processed by the Bottleneck Pr…
Figure 5
Figure 5. Figure 5: Illustration of the Crossmodal Amplifier within the branch of Modality A. The module is primarily composed of an Inverted Bottleneck Projection. Encoder features F B E i , i = {1, 2, 3} from the other modality (Modality B) are first processed by this module to generate…
Figure 6
Figure 6. Figure 6: Inference process of our proposed method. The feature evolution visualization highlights that anomalies in each modality are effectively cap￾tured. In resulting anomaly maps, darker gray regions indicate higher anomaly scores. proach in RD, we optimize the features out…
Figure 7
Figure 7. Figure 7: Performance comparison of feature learning [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 9
Figure 9. Figure 9: Visualization of anomaly maps of different modality branches on [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]
Figure 8
Figure 8. Figure 8: Visualization of anomaly maps of different modality branches on [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 10
Figure 10. Figure 10: Qualitative ablation results of Crossmodal Tuners on MVTec 3D-AD. [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]
Figure 11
Figure 11. Figure 11: Visualization of the features output by CF and CA. [PITH_FULL_IMAGE:figures/full_fig_p009_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 27 canonical work pages

  1. [1]

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

    P. Bergmann, M. Fauser, D. Sattlegger, and C. 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, 2019, pp. 9592–9600

    work page 2019

  2. [2]

    Be- yond dents and scratches: Logical constraints in unsupervised anomaly detection and localization,

    P. Bergmann, K. Batzner, M. Fauser, D. Sattlegger, and C. Steger, “Be- yond dents and scratches: Logical constraints in unsupervised anomaly detection and localization,” International Journal of Computer Vision , vol. 130, no. 4, pp. 947–969, 2022

    2022

  3. [3]

    Spot-the- difference self-supervised pre-training for anomaly detection and seg- mentation,

    Y . Zou, J. Jeong, L. Pemula, D. Zhang, and O. Dabeer, “Spot-the- difference self-supervised pre-training for anomaly detection and seg- mentation,” in European Conference on Computer Vision . Springer, 2022, pp. 392–408

    2022

  4. [4]

    The mvtec 3d-ad dataset for unsupervised 3d anomaly detection and localization,

    P. Bergmann, X. Jin, D. Sattlegger, and C. Steger, “The mvtec 3d-ad dataset for unsupervised 3d anomaly detection and localization,” arXiv preprint arXiv:2112.09045, 2021

    arXiv 2021

  5. [5]

    Asymmetric student-teacher networks for industrial anomaly detection,

    M. Rudolph, T. Wehrbein, B. Rosenhahn, and B. Wandt, “Asymmetric student-teacher networks for industrial anomaly detection,” in Proceed- ings of the IEEE/CVF winter conference on applications of computer vision, 2023, pp. 2592–2602

    work page 2023

  6. [6]

    Mul- timodal industrial anomaly detection by crossmodal feature mapping,

    A. Costanzino, P. Z. Ramirez, G. Lisanti, and L. Di Stefano, “Mul- timodal industrial anomaly detection by crossmodal feature mapping,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 17 234–17 243

    work page 2024

  7. [7]

    Back to the feature: classical 3d features are (almost) all you need for 3d anomaly detection,

    E. Horwitz and Y . Hoshen, “Back to the feature: classical 3d features are (almost) all you need for 3d anomaly detection,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 2968–2977

    work page 2023

  8. [8]

    Multimodal industrial anomaly detection via hybrid fusion,

    Y . Wang, J. Peng, J. Zhang, R. Yi, Y . Wang, and C. Wang, “Multimodal industrial anomaly detection via hybrid fusion,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023, pp. 8032–8041

    work page 2023

Show all 38 references
  1. [9]

    Rethinking reverse distillation for multi-modal anomaly detection,

    Z. Gu, J. Zhang, L. Liu, X. Chen, J. Peng, Z. Gan, G. Jiang, A. Shu, Y . Wang, and L. Ma, “Rethinking reverse distillation for multi-modal anomaly detection,” in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 38, no. 8, 2024, pp. 8445–8453

    2024

  2. [10]

    Anomaly detection via reverse distillation from one-class embedding,

    H. Deng and X. Li, “Anomaly detection via reverse distillation from one-class embedding,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2022, pp. 9737–9746

    2022

  3. [11]

    Ganomaly: Semi- supervised anomaly detection via adversarial training,

    S. Akcay, A. Atapour-Abarghouei, and T. P. Breckon, “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 . Spr...

    2018

  4. [12]

    Attribute restoration framework for anomaly detection,

    F. Ye, C. Huang, J. Cao, M. Li, Y . Zhang, and C. Lu, “Attribute restoration framework for anomaly detection,” IEEE Transactions on Multimedia, vol. 24, pp. 116–127, 2020

    2020

  5. [13]

    A unified model for multi-class anomaly detection,

    Z. You, L. Cui, Y . Shen, K. Yang, X. Lu, Y . Zheng, and X. Le, “A unified model for multi-class anomaly detection,” Advances in Neural Information Processing Systems , vol. 35, pp. 4571–4584, 2022

    2022

  6. [14]

    Self-supervised masking for unsupervised anomaly detection and localization,

    C. Huang, Q. Xu, Y . Wang, Y . Wang, and Y . Zhang, “Self-supervised masking for unsupervised anomaly detection and localization,” IEEE Transactions on Multimedia , vol. 25, pp. 4426–4438, 2022

    2022

  7. [15]

    Unsuper- vised surface anomaly detection with diffusion probabilistic model,

    X. Zhang, N. Li, J. Li, T. Dai, Y . Jiang, and S.-T. Xia, “Unsuper- vised surface anomaly detection with diffusion probabilistic model,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 6782–6791

    2023

  8. [16]

    Feature reconstruc- tion with disruption for unsupervised video anomaly detection,

    C. Tao, C. Wang, S. Lin, S. Cai, D. Li, and J. Qian, “Feature reconstruc- tion with disruption for unsupervised video anomaly detection,” IEEE Transactions on Multimedia , vol. 26, pp. 10 160–10 173, 2024

    2024

  9. [17]

    A diffusion-based framework for multi-class anomaly detection,

    H. He, J. Zhang, H. Chen, X. Chen, Z. Li, X. Chen, Y . Wang, C. Wang, and L. Xie, “A diffusion-based framework for multi-class anomaly detection,” in Proceedings of the AAAI conference on artificial intelligence, vol. 38, no. 8, 2024, pp. 8472–8480

    2024

  10. [18]

    Towards total recall in industrial anomaly detection,

    K. Roth, L. Pemula, J. Zepeda, B. Sch ¨olkopf, T. Brox, and P. Gehler, “Towards total recall in industrial anomaly detection,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 14 318–14 328

    2022

  11. [19]

    Pni: industrial anomaly detection using position and neighborhood information,

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

    2023

  12. [20]

    Target be- fore shooting: Accurate anomaly detection and localization under one millisecond via cascade patch retrieval,

    H. Li, J. Hu, B. Li, H. Chen, Y . Zheng, and C. Shen, “Target be- fore shooting: Accurate anomaly detection and localization under one millisecond via cascade patch retrieval,” IEEE Transactions on Image Processing, 2024

    2024

  13. [21]

    Cutpaste: Self-supervised learning for anomaly detection and localization,

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

    2021

  14. [22]

    Draem-a discriminatively trained reconstruction embedding for surface anomaly detection,

    V . Zavrtanik, M. Kristan, and D. Sko ˇcaj, “Draem-a discriminatively trained reconstruction embedding for surface anomaly detection,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 8330–8339

    2021

  15. [23]

    A comprehensive augmentation framework for anomaly detection,

    J. Lin and Y . Yan, “A comprehensive augmentation framework for anomaly detection,” in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 38, no. 8, 2024, pp. 8742–8749

    2024

  16. [24]

    Multiresolution knowledge distillation for anomaly detection,

    M. Salehi, N. Sadjadi, S. Baselizadeh, M. H. Rohban, and H. R. Rabiee, “Multiresolution knowledge distillation for anomaly detection,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 14 902–14 912

    2021

  17. [25]

    Revisiting reverse distillation for anomaly detection,

    T. D. Tien, A. T. Nguyen, N. H. Tran, T. D. Huy, S. Duong, C. D. T. Nguyen, and S. Q. Truong, “Revisiting reverse distillation for anomaly detection,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2023, pp. 24 511–24 520

    2023

  18. [26]

    Template-guided hierarchical feature restoration for anomaly detection,

    H. Guo, L. Ren, J. Fu, Y . Wang, Z. Zhang, C. Lan, H. Wang, and X. Hou, “Template-guided hierarchical feature restoration for anomaly detection,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023, pp. 6447–6458

    2023

  19. [27]

    Remembering normality: Memory-guided knowl- edge distillation for unsupervised anomaly detection,

    Z. Gu, L. Liu, X. Chen, R. Yi, J. Zhang, Y . Wang, C. Wang, A. Shu, G. Jiang, and L. Ma, “Remembering normality: Memory-guided knowl- edge distillation for unsupervised anomaly detection,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023, pp. 1...

    2023

  20. [28]

    Contextual affinity distillation for image anomaly detection,

    J. Zhang, M. Suganuma, and T. Okatani, “Contextual affinity distillation for image anomaly detection,” in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , 2024, pp. 149–158

    2024

  21. [29]

    Recontrast: Domain-specific anomaly detection via contrastive reconstruction,

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

    2024

  22. [30]

    The eyecandies dataset for unsupervised multimodal anomaly detection and localization,

    L. Bonfiglioli, M. Toschi, D. Silvestri, N. Fioraio, and D. De Gregorio, “The eyecandies dataset for unsupervised multimodal anomaly detection and localization,” in Proceedings of the Asian Conference on Computer Vision, 2022, pp. 3586–3602

    2022

  23. [31]

    Real3d-ad: A dataset of point cloud anomaly detection,

    J. Liu, G. Xie, R. Chen, X. Li, J. Wang, Y . Liu, C. Wang, and F. Zheng, “Real3d-ad: A dataset of point cloud anomaly detection,” Advances in Neural Information Processing Systems , vol. 36, 2024

    2024

  24. [32]

    Shape- guided dual-memory learning for 3d anomaly detection,

    Y .-M. Chu, L. Chieh, T.-I. Hsieh, H.-T. Chen, and T.-L. Liu, “Shape- guided dual-memory learning for 3d anomaly detection,” 2023

    2023

  25. [33]

    Self-supervised feature adaptation for 3d industrial anomaly detection,

    Y . Tu, B. Zhang, L. Liu, Y . Li, J. Zhang, Y . Wang, C. Wang, and C. Zhao, “Self-supervised feature adaptation for 3d industrial anomaly detection,” in European Conference on Computer Vision . Springer, 2024, pp. 75– 91

    2024

  26. [34]

    Easynet: An easy network for 3d industrial anomaly detection,

    R. Chen, G. Xie, J. Liu, J. Wang, Z. Luo, J. Wang, and F. Zheng, “Easynet: An easy network for 3d industrial anomaly detection,” in Proceedings of the 31st ACM International Conference on Multimedia , 2023, pp. 7038–7046

    2023

  27. [35]

    Cheating depth: Enhancing 3d surface anomaly detection via depth simulation,

    V . Zavrtanik, M. Kristan, and D. Sko ˇcaj, “Cheating depth: Enhancing 3d surface anomaly detection via depth simulation,” in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , 2024, pp. 2164–2172

    2024

  28. [36]

    Keep dræming: Discriminative 3d anomaly detection through anomaly simulation,

    ——, “Keep dræming: Discriminative 3d anomaly detection through anomaly simulation,” Pattern Recognition Letters , vol. 181, pp. 113– 119, 2024

    2024

  29. [37]

    Anomaly detection in 3d point clouds using deep geometric descriptors,

    P. Bergmann and D. Sattlegger, “Anomaly detection in 3d point clouds using deep geometric descriptors,” in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2023, pp. 2613– 2623

    2023

  30. [38]

    Wide residual networks,

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

    2016 arXiv

Pith tools

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