Pith. sign in

REVIEW 2 major objections 6 minor 46 references

SiM3D: Single-instance Multiview Multimodal and Multisetup 3D Anomaly Detection Benchmark

T0 review · 2 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read This paper introduces SiM3D, a benchmark that asks anomaly-detection systems to fuse multiview multimodal scans into a voxel-level 3D Anomaly Volume while training from a single real or synthetic object.

desk verdict A genuinely useful multiview 3D anomaly-detection benchmark whose headline CAD-to-real claim is contradicted by its own Section 3 — fixable, and worth refereeing. read the letter →

arxiv 2506.21549 v2 pith:RKWZINTF submitted 2025-06-26 cs.CV

Alex Costanzino , Pierluigi Zama Ramirez , Luigi Lella , Matteo Ragaglia , Alessandro Oliva , Giuseppe Lisanti , Luigi Di Stefano This is my paper
classification cs.CV
keywords anomalydetection3Dsegmentationmultiviewbenchmarkmultimodalfusionsynthetic-to-realVolumesingle-instancelearningindustrialinspection
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

SiM3D is a dataset and benchmark for industrial anomaly detection and segmentation in 3D. The paper argues that existing benchmarks score 2D anomaly maps from single viewpoints, whereas manufacturing would benefit from a voxel-based Anomaly Volume built by integrating many views plus image and 3D data. To move the field in that direction, the authors collected 360-degree scans of 333 instances of eight manufactured object types, each with 12 Mpx greyscale images, point clouds of 5–7 million points, integrated meshes, a per-type CAD model, and manually refined 3D ground truths. The benchmark defines two single-instance training setups, real-to-real and synthetic-to-real, sharing one test set, and claims to be the first ADS benchmark to evaluate generalization from synthetic training data to real test data. Widely used single-view methods score far below saturation when extended to this task, which the paper reads as evidence that the task is new, difficult, and open.

What carries the argument

The central object is the Anomaly Volume, a voxel grid in which each voxel carries a defect-likelihood score, with ground truth defined on the same 2 mm voxel grid and metrics extended to that representation: instance-level AUROC (I-AUROC) for detection and voxel-level AUPRO (V-AUPRO), integrated to a 1% false-positive rate, for segmentation. To turn single-view methods into multiview baselines, the paper projects each view's 2D anomaly map onto the integrated mesh using calibrated camera intrinsics and known view poses, accumulates per-voxel scores, and takes the maximum across views. The dataset itself is built on an acquisition pipeline: an industrial scanner mounted on a robot arm collects 12–36 views per object with known poses, and synthetic training scans are rendered in Blender from a real instance's mesh aligned to the CAD model, using the same camera intrinsics and viewpoints as the real captures.

What would settle it

A decisive check is to render the synthetic training views twice, once from the CAD geometry alone and once from the released reference-mesh geometry, then compare each set to the real test scans using distribution-shift measures such as FID for images or Chamfer distance for point clouds. If the CAD-only renders sit much farther from the real test data than the reference-mesh renders do, the synthetic-to-real benchmark overstates generalization from a pure CAD prototype; a companion experiment rerunning PatchCore on CAD-only training and comparing its detection and segmentation scores would quantify how much of the reported synthetic performance relies on real geometry.

Watch

Extended reading notes

Core claim

The central claim is that a meaningful 3D anomaly-detection benchmark should output an Anomaly Volume, a voxel grid whose cells carry defect likelihoods, rather than a 2D anomaly map, and that producing such a volume requires integrating multiview and multimodal observations at test time. SiM3D provides the evaluation ground: 333 instances across eight manufactured objects, 12 to 36 views per instance, greyscale images and high-density point clouds from an industrial scanner, integrated meshes, CAD models for each type, and voxel-level ground truths in which every defect carries its own ID. The paper further claims that a single nominal instance, whether real or synthetic, can serve as the entire training set, and that training from CAD-derived scans and testing on real scans is a setup no previous ADS benchmark covers. Adapted baselines built from PatchCore, EfficientAD, BTF, M3DM, CFM, and AST perform well below the usual saturation levels of 2D benchmarks, and RGB-only methods often beat multimodal ones when point clouds are at their native high resolution, which the authors attribute to the lack of 3D backbones that scale to millions of points.

Load-bearing premise

The load-bearing premise is that the synthetic training scans truly represent the CAD prototype; in fact, they are rendered from a real instance's mesh aligned to the CAD model, so the claimed synthetic-to-real generalization may be tested on a narrower domain gap than a pure CAD-to-real deployment.

Editorial extensions

If this is right

  • A method that truly solves SiM3D will need to consume all views jointly instead of aggregating per-view 2D anomaly maps; max-pooling projected scores is only a first approximation.
  • Because the same test set serves both real-to-real and synthetic-to-real, the benchmark yields a direct, per-object measure of the cost of training from synthetic data alone.
  • Multimodal methods designed for point clouds of about 8k points do not scale to SiM3D's millions of points; depth maps act as a workable intermediate representation until high-resolution 3D backbones exist.
  • The task's difficulty comes from the 3D multiview setup rather than from grayscale input, since PatchCore on grayscale MVTec AD matches its official published results while dropping substantially on SiM3D.
  • The per-defect IDs in the voxel ground truths make V-AUPRO able to reward methods that localize each defect as a connected region, not merely those that overlap the anomaly mask somewhere.

Reading between the lines

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

  • Editorial extension: if the synthetic training scans are rendered from a real reference mesh aligned to the CAD model, the synthetic-to-real track may measure a smaller domain gap than a pure CAD-to-real deployment; a stronger test would render from the CAD geometry alone.
  • Editorial extension: the dataset's defect-size distribution, dominated by small anomalies, invites a follow-up analysis of whether baseline failures track defect volume, defect type (2D, 3D, or multimodal), or both.
  • Editorial extension: the Anomaly Volume representation could feed directly into robot-guided rework or closed-loop quality control; benchmarks could add action-oriented evaluations such as the accuracy of a gripper targeting the top-scoring voxel cluster.
  • Editorial extension: the observed gap between depth-map and point-cloud baselines suggests a testable hypothesis that a foundation model trained specifically on high-resolution depth or point-cloud data would close much of the multimodal gap; SiM3D would be the natural testbed.
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

2 major / 6 minor

Summary. SiM3D is a new benchmark for multiview, multimodal 3D anomaly detection and segmentation. The paper introduces a dataset of 333 instances of 8 manufactured object types, each with high-resolution greyscale images, point clouds and integrated meshes acquired by a calibrated industrial sensor/robot setup, plus a CAD model per object type. The benchmark defines the task of producing a voxel-based anomaly volume from multiview and multimodal inputs, provides manual 3D voxel ground truths, and introduces I-AUROC and V-AUPRO metrics adapted to this setting. The authors adapt six established single-view anomaly-detection methods to the new task and evaluate them in two setups: real2real (one real nominal instance for training) and synth2real (rendered synthetic data for training, with the same real test set). The reported baseline performance is relatively low, suggesting the task is challenging and not yet solved.

Significance. If the claims are substantiated, SiM3D could become a useful evaluation standard for 3D anomaly detection, filling a clear gap: existing benchmarks mostly output 2D anomaly maps or use single-view test inputs, whereas SiM3D explicitly targets multiview integration into a voxel-based anomaly volume and evaluates synthetic-to-real transfer. The dataset construction is described in considerable detail: sensor calibration, view planning, 3D ground-truth labelling with manual refinement, and baseline adaptations from official code. The authors also provide sanity checks (e.g., PatchCore on grayscale MVTec AD, and a single-view vs. multiview comparison). The low baseline scores and open challenges identified (high-resolution point cloud processing, single-instance training, domain shift) are plausible and likely to stimulate follow-up work. These strengths make the benchmark potentially valuable, provided that the synthetic data issue and the inconsistencies in the reported split statistics are resolved.

major comments (2)
  1. [Sec. 3, 'Synthetic data generation' (and Fig. 3)] The synthetic training scans are rendered from a reference mesh selected from the collected real instances, aligned to the prototype CAD model, rather than from the CAD geometry itself. This contradicts the abstract ('we deploy the available CAD model') and Sec. 1 ('we deploy the available CAD model to obtain the same kind of data'), and it weakens the headline claim of being the first benchmark for generalization from synthetic training data to real test data if that claim is intended to be about CAD-prototype-to-physical-object transfer. As described, the setup is closer to rendered-real-to-real; the CAD model is used only as an alignment target. Please either modify the rendering pipeline so that the CAD mesh is the actual geometric source, or revise the claims and terminology throughout (including 'synthetic prototype (e.g., an object's CAD model)') to accurately describe what is evaluated.
  2. [Table 2] For every object type, the total instance count does not equal the sum of the listed train and test counts. For example, Plastic Stool has 22 total but 1 (train) + 10 (nominal test) + 10 (anomalous test) = 21; Rubbish Bin has 42 total but 41 accounted for; and the same one-over count appears for all eight types. This suggests an undocumented held-out instance per type, likely the reference mesh used for synthetic generation, is excluded from the train/test split without being stated. Please explicitly specify how the reference mesh instance is treated relative to the benchmark splits and correct the table so that the composition of the dataset is fully reproducible.
minor comments (6)
  1. [Sec. 3, first paragraph] The statement that 'only half of the instances within each object type are modified' is not supported by the counts in Table 2 / Table B; for instance, Plastic Stool has 10 anomalous instances out of 22 total. Please state the exact number of anomalous instances per type or rephrase the sentence.
  2. [Sec. 5, first paragraph] The claim that PatchCore 'can achieve the best detection in both setups' is not supported by Table 3: in the synth2real setup, AST has a higher mean I-AUROC (0.544) than PatchCore with DINO-v2 (0.540) or with WRN-101 (0.451). Please correct the summary of the results.
  3. [Supplement S.7] The FID-based analysis in Table C is informative for understanding the synth2real domain gap; I recommend referencing it in the main text (e.g., in Sec. 5) rather than only in the supplement.
  4. [Sec. 4.1, V-AUPRO definition] Please clarify whether the false-positive rate is computed per object or over the whole dataset, and define 'non-empty voxels' (e.g., voxels that intersect the reconstructed mesh). This would make the metric fully reproducible.
  5. [Abstract and Table 1] The abstract describes point clouds of ~7M points, while Sec. 3 reports the sensor can output up to 12M points and Table 1 lists 5M–7M; please make the numbers consistent.
  6. [Figure 1 and Fig. 3] The synthetic training sample is described as coming from a CAD model, but according to Sec. 3 it is rendered from a real reference mesh; please align the figure captions with the actual pipeline.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the benchmark evaluates external methods on a new dataset, and the only self-citation is non-load-bearing, while the synthetic-data provenance issue is a correctness concern rather than a circular derivation.

full rationale

The paper's central contribution is a new dataset, task definition, and evaluation protocol for multiview, multimodal, voxel-based 3D anomaly detection with single-instance training. The baselines are adapted from a range of external methods, including PatchCore, EfficientAD, BTF, M3DM, and AST, and are evaluated on held-out real test scans using standard adapted metrics. No fitted parameter is renamed as a prediction, no uniqueness theorem is imported from the authors' prior work, and no ansatz is smuggled in via citation. The authors do cite their own CFM paper [14] for the choice of the 1% AUPRO integration bound and include CFM as one baseline, but these self-citations are not load-bearing: the benchmark's validity and the paper's conclusions do not depend on CFM's published results, and CFM actually performs poorly in the reported experiments. A separate, non-circularity concern is that Section 3 states the 'synthetic' training renders are produced by selecting 'a reference mesh from the collected real instances' and aligning it to the CAD model, rather than rendering the CAD geometry itself; this weakens the claimed CAD-to-real generalization setup and is a construct-validity or correctness risk, but it is not a circular derivation because the benchmark still evaluates methods on held-out real scans and does not reduce any prediction to its own inputs.

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

The benchmark rests on domain assumptions about single-instance sufficiency and calibration accuracy, and on the contested assumption that the synthetic data are CAD-derived. No new physical entities are introduced, and no numbers are fitted to make a derivation work.

assumptions (4)
  • domain assumption A single nominal instance is sufficient to represent the nominal distribution for each object class
    This underpins the single-instance training setup in both real2real and synth2real; if one object cannot represent the class, the benchmark's premise fails.
  • domain assumption The integrated mesh and per-view point clouds are accurately calibrated so that 2D projections align with 3D geometry
    Baselines project 2D anomaly maps into voxels via these calibrations; calibration errors would directly affect V-AUPRO.
  • domain assumption The voxel size of 2 mm is small enough to capture all relevant defects
    Ground truth and Anomaly Volumes are quantized at 2 mm; any defect smaller than this would be missed or underestimated.
  • ad hoc to paper The synthetic training data is generated from the CAD model
    The abstract and intro claim CAD-based rendering, but Sec. 3 says a real reference mesh is aligned to the CAD and rendered. This inconsistency is load-bearing for the synth2real claim.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SiM3D: Single-instance Multiview Multimodal and Multisetup 3D Anomaly Detection Benchmark." pith.science (2026). https://pith.science/paper/RKWZINTF

@misc{pith2026250621549,
  author       = {Pith},
  title        = {Pith review of: SiM3D: Single-instance Multiview Multimodal and Multisetup 3D Anomaly Detection Benchmark},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RKWZINTF}},
  note         = {Machine review of arXiv:2506.21549}
}
read the original abstract

We propose SiM3D, the first benchmark considering the integration of multiview and multimodal information for comprehensive 3D anomaly detection and segmentation (ADS), where the task is to produce a voxel-based Anomaly Volume. Moreover, SiM3D focuses on a scenario of high interest in manufacturing: single-instance anomaly detection, where only one object, either real or synthetic, is available for training. In this respect, SiM3D stands out as the first ADS benchmark that addresses the challenge of generalising from synthetic training data to real test data. SiM3D includes a novel multimodal multiview dataset acquired using top-tier industrial sensors and robots. The dataset features multiview high-resolution images (12 Mpx) and point clouds (7M points) for 333 instances of eight types of objects, alongside a CAD model for each type. We also provide manually annotated 3D segmentation GTs for anomalous test samples. To establish reference baselines for the proposed multiview 3D ADS task, we adapt prominent singleview methods and assess their performance using novel metrics that operate on Anomaly Volumes.

Figures

Figures reproduced from arXiv: 2506.21549 by the authors.

Figure 1
Figure 1. SiM3D dataset overview. From top to bottom: the single-instance real and synthetic training samples for object Plas￾tic Stool, one of the anomalous samples from the test set. MVTec 3D-AD and [6] and Eyecandies [7] introduced the first benchmarks deploying multimodal inputs to enhance ADS performance. Then, PAD [45] and Real-IAD [35] introduced novel benchmarks that leverage multiple RGB views of objects. Although th… view at source ↗
Figure 2
Figure 2. 2D vs. 3D anomaly detection and segmentation. Typ￾ical ADS benchmarks rely on 2D Anomaly Maps and 2D ground￾truths (left) to assess segmentation performance. SiM3D proposes to do so by 3D Anomaly Volumes and 3D ground-truths (right). erate Anomaly Maps, i.e., images that encode the likeli￾hood of each pixel belonging to a defect, as shown in [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. SiM3D dataset creation pipeline. (1) We purchased eight types of objects with different properties, each associated with a CAD model. We altered part of the samples to create realistic anomalies; (2) We designed and calibrated an acquisition setup consisting of a high-precision 3D sensor with high-resolution cameras on top an industrial robot arm; (3) We acquired 360-degree scans of each object instance to populate … view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Extending singleview methods. Anomaly Maps ob￾tained by processing singleview inputs with standard ADS meth￾ods are projected into the 3D and aggregated into a voxel grid, with each voxel containing a list of scores from multiple views. The Anomaly Volume is obtained b…
Figure 5
Figure 5. Figure 5: Qualitatives. Selected views with defects (left) and Anomaly Volumes for several methods (right) downsampled for visualisation. Method Modality real2real synth2real Pl. Stool Rub. Bin W. Vase B. Furn. Cont. Pl. Vase W. Stool Sink Cab. Mean Pl. Stool Rub. Bin W. Vase B.…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

46 extracted references · 39 canonical work pages

  1. [1]

    Efficien- tad: Accurate visual anomaly detection at millisecond-level latencies

    Kilian Batzner, Lars Heckler, and Rebecca K ¨onig. Efficien- tad: Accurate visual anomaly detection at millisecond-level latencies. In Proceedings of the IEEE/CVF Winter Con- ference on Applications of Computer Vision (WACV), pages 128–138, 2024. 3, 6, 13

    work page 2024

  2. [2]

    Deep nearest neighbor anomaly detection

    Liron Bergman, Niv Cohen, and Yedid Hoshen. Deep nearest neighbor anomaly detection. arXiv preprint arXiv:2002.10445, 2020. 3

    arXiv 2002

  3. [3]

    Improving unsupervised defect seg- mentation by applying structural similarity to autoencoders

    Paul Bergmann, Sindy L ¨owe, Michael Fauser, David Sattleg- ger, and Carsten Steger. Improving unsupervised defect seg- mentation by applying structural similarity to autoencoders. arXiv preprint arXiv:1807.02011, 2018. 3

    arXiv 2018

  4. [4]

    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 Pat- tern Recognition (CVPR), 2019. 1, 3, 6

    work page 2019

  5. [5]

    Beyond dents and scratches: Logical constraints in unsupervised anomaly detection and localization

    Paul Bergmann, Kilian Batzner, Michael Fauser, David Sat- tlegger, and Carsten Steger. Beyond dents and scratches: Logical constraints in unsupervised anomaly detection and localization. International Journal of Computer Vision, 130,

    work page

  6. [6]

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

    Paul Bergmann, Jin Xin, David Sattlegger, and Carsten Ste- ger. The mvtec 3d-ad dataset for unsupervised 3d anomaly detection and localization. In Proceedings of the 17th Inter- national Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications , pages 202– 213, 2022. 1, 2, 3, 7

    work page 2022

  7. [7]

    The eyecandies dataset for unsupervised multimodal anomaly detection and local- ization

    Luca Bonfiglioli, Marco Toschi, Davide Silvestri, Nicola Fioraio, and Daniele De Gregorio. The eyecandies dataset for unsupervised multimodal anomaly detection and local- ization. In Proceedings of the 16th Asian Conference on Computer Vision (ACCV2022), 2022. ACCV . 1, 2, 3, 7

    work page 2022

  8. [8]

    Emerg- ing properties in self-supervised vision transformers

    Mathilde Caron, Hugo Touvron, Ishan Misra, Herv ´e J´egou, Julien Mairal, Piotr Bojanowski, and Armand Joulin. Emerg- ing properties in self-supervised vision transformers. In Pro- ceedings of the International Conference on Computer Vi- sion (ICCV), 2021. 3

    2021

Show all 46 references
  1. [9]

    Self-supervised normal- izing flows for image anomaly detection and localization

    Li-Ling Chiu and Shang-Hong Lai. Self-supervised normal- izing flows for image anomaly detection and localization. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 2926–2935, 2023. 3

    2023

  2. [10]

    Sub-image anomaly detection with deep pyramid correspondences

    Niv Cohen and Yedid Hoshen. Sub-image anomaly detection with deep pyramid correspondences. ArXiv, 2020. 3

    2020

  3. [11]

    Blender - a 3D modelling and rendering package

    Blender Online Community. Blender - a 3D modelling and rendering package. Blender Foundation, Stichting Blender Foundation, Amsterdam, 2024. 5

    2024

  4. [12]

    CloudCompare - 3D point cloud and mesh processing software Open Source Project

    CloudCompare Online Community. CloudCompare - 3D point cloud and mesh processing software Open Source Project. EDF, 2024. 5

    2024

  5. [13]

    Test time training for industrial anomaly seg- mentation

    Alex Costanzino, Pierluigi Zama Ramirez, Mirko Del Moro, Agostino Aiezzo, Giuseppe Lisanti, Samuele Salti, and Luigi Di Stefano. Test time training for industrial anomaly seg- mentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR...

    2024

  6. [14]

    Multimodal industrial anomaly detec- tion by crossmodal feature mapping

    Alex Costanzino, Pierluigi Zama Ramirez, Giuseppe Lisanti, and Luigi Di Stefano. Multimodal industrial anomaly detec- tion by crossmodal feature mapping. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recogni- tion, 2024. CVPR. 1, 3, 6, 14

    2024

  7. [15]

    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 Inter- national Conference on Pattern Recognition, pages 475–489. Springer, 2021. 3

    2021

  8. [16]

    Cflow-ad: Real-time unsupervised anomaly detection with localization via conditional normalizing flows

    Denis Gudovskiy, Shun Ishizaka, and Kazuki Kozuka. Cflow-ad: Real-time unsupervised anomaly detection with localization via conditional normalizing flows. In Proceed- ings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 98–107, 2022. 3

    2022

  9. [17]

    Masked autoencoders are scalable vision learners

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Doll´ar, and Ross Girshick. Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 16000– 16009, 2022. 3, 6

    2022

  10. [18]

    Back to the feature: clas- sical 3d features are (almost) all you need for 3d anomaly detection

    Eliahu Horwitz and Yedid Hoshen. Back to the feature: clas- sical 3d features are (almost) all you need for 3d anomaly detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 2967– 2976, 2023. 3, 6, 13

    2023

  11. [19]

    Divide-and-assemble: Learning block-wise memory for unsupervised anomaly detection

    Jinlei Hou, Yingying Zhang, Qiaoyong Zhong, Di Xie, Shil- iang Pu, and Hong Zhou. Divide-and-assemble: Learning block-wise memory for unsupervised anomaly detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 8791–8800, 2021. 3

    2021

  12. [20]

    Cutpaste: Self-supervised learning for anomaly de- tection and localization

    Chun-Liang Li, Kihyuk Sohn, Jinsung Yoon, and Tomas Pfister. Cutpaste: Self-supervised learning for anomaly de- tection and localization. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 9664–9674, 2021. 3

    2021

  13. [21]

    Real3d-ad: A dataset of point cloud anomaly detection

    Jiaqi Liu, Guoyang Xie, Xinpeng Li, Jinbao Wang, Yong Liu, Chengjie Wang, Feng Zheng, et al. Real3d-ad: A dataset of point cloud anomaly detection. In Thirty-seventh Con- ference on Neural Information Processing Systems Datasets and Benchmarks Track, 2023. 3

    2023

  14. [22]

    Deep in- dustrial image anomaly detection: A survey

    Jiaqi Liu, Guoyang Xie, Jingbao Wang, Shangnian Li, Chengjie Wang, Feng Zheng, and Yaochu Jin. Deep in- dustrial image anomaly detection: A survey. arXiv preprint arXiv:2301.11514, 2, 2023. 3

    2023 arXiv

  15. [23]

    Mocca: Multilayer one-class classification for anomaly de- tection

    Fabio Valerio Massoli, Fabrizio Falchi, Alperen Kantarci, S ¸eymanur Akti, Hazim Kemal Ekenel, and Giuseppe Amato. Mocca: Multilayer one-class classification for anomaly de- tection. IEEE Transactions on Neural Networks and Learn- ing Systems, 33(6):2313–2323, 2021. 3

    2021

  16. [24]

    Maxime Oquab, Timoth ´ee Darcet, Theo Moutakanni, Huy V . V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, Rus- sell Howes, Po-Yao Huang, Hu Xu, Vasu Sharma, Shang- Wen Li, Wojciech Galuba, Mike Rabbat, Mido Assran, Ni...

    2023

  17. [25]

    Masked autoencoders for point cloud self-supervised learning

    Yatian Pang, Wenxiao Wang, Francis EH Tay, Wei Liu, Yonghong Tian, and Li Yuan. Masked autoencoders for point cloud self-supervised learning. In Computer Vision–ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part II, pages 604–621. Springer,

    2022

  18. [26]

    Inpainting transformer for anomaly detection

    Jonathan Pirnay and Keng Chai. Inpainting transformer for anomaly detection. In International Conference on Image Analysis and Processing, pages 394–406. Springer, 2022. 3

    2022

  19. [27]

    Panda: Adapting pretrained features for anomaly detection and segmentation

    Tal Reiss, Niv Cohen, Liron Bergman, and Yedid Hoshen. Panda: Adapting pretrained features for anomaly detection and segmentation. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 2806–2814, 2021. 3

    2021

  20. [28]

    Model- ing the distribution of normal data in pre-trained deep fea- tures for anomaly detection

    Oliver Rippel, Patrick Mertens, and Dorit Merhof. Model- ing the distribution of normal data in pre-trained deep fea- tures for anomaly detection. In 2020 25th International Con- ference on Pattern Recognition (ICPR) , pages 6726–6733. IEEE, 2021

    2020

  21. [29]

    Self-supervised predictive convo- lutional attentive block for anomaly detection

    Nicolae-C ˘at˘alin Ristea, Neelu Madan, Radu Tudor Ionescu, Kamal Nasrollahi, Fahad Shahbaz Khan, Thomas B Moes- lund, and Mubarak Shah. Self-supervised predictive convo- lutional attentive block for anomaly detection. In Proceed- ings of the IEEE/CVF conference on computer vi...

    2022

  22. [30]

    Towards total re- call in industrial anomaly detection

    Karsten Roth, Latha Pemula, Joaquin Zepeda, Bernhard Sch¨olkopf, Thomas Brox, and Peter Gehler. Towards total re- call in industrial anomaly detection. In Proceedings of 2022 IEEE Conference on Computer Vision and Pattern Recogni- tion, pages 14298–14308, 2022. 1, 3, 6, 13, 16

    2022

  23. [31]

    Same same but differnet: Semi-supervised defect detection with normalizing flows

    Marco Rudolph, Bastian Wandt, and Bodo Rosenhahn. Same same but differnet: Semi-supervised defect detection with normalizing flows. In Proceedings of the IEEE/CVF winter conference on applications of computer vision, pages 1907– 1916, 2021. 3

    1907

  24. [32]

    Asymmetric student-teacher networks for indus- trial anomaly detection

    Marco Rudolph, Tom Wehrbein, Bodo Rosenhahn, and Bas- tian Wandt. Asymmetric student-teacher networks for indus- trial anomaly detection. In Winter Conference on Applica- tions of Computer Vision (WACV), 2023. 3, 14

    2023

  25. [33]

    Fast point feature histograms (fpfh) for 3d registration

    Radu Bogdan Rusu, Nico Blodow, and Michael Beetz. Fast point feature histograms (fpfh) for 3d registration. In 2009 IEEE International Conference on Robotics and Automation, pages 3212–3217, 2009. 3, 6

    2009

  26. [34]

    Learning and evaluating representations for deep one-class classification

    Kihyuk Sohn, Chun-Liang Li, Jinsung Yoon, Minho Jin, and Tomas Pfister. Learning and evaluating representations for deep one-class classification. In International Conference on Learning Representations, 2021. 3

    2021

  27. [35]

    Real-iad: A real-world multi-view dataset for benchmarking versatile industrial anomaly detec- tion

    Chengjie Wang, Wenbing Zhu, Bin-Bin Gao, Zhenye Gan, Jiangning Zhang, Zhihao Gu, Shuguang Qian, Mingang Chen, and Lizhuang Ma. Real-iad: A real-world multi-view dataset for benchmarking versatile industrial anomaly detec- tion. In Proceedings of the IEEE/CVF Conference on Com-...

    2024

  28. [36]

    Multimodal industrial anomaly detection via hybrid fusion

    Yue Wang, Jinlong Peng, Jiangning Zhang, Ran Yi, Yabiao Wang, and Chengjie Wang. Multimodal industrial anomaly detection via hybrid fusion. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 8032–8041, 2023. 1, 3, 6, 14

    2023

  29. [37]

    Anoddpm: Anomaly detection with de- noising diffusion probabilistic models using simplex noise

    Julian Wyatt, Adam Leach, Sebastian M Schmon, and Chris G Willcocks. Anoddpm: Anomaly detection with de- noising diffusion probabilistic models using simplex noise. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 650–656, 2022. 3

    2022

  30. [38]

    Memseg: A semi- supervised method for image surface defect detection using differences and commonalities

    Minghui Yang, Peng Wu, and Hui Feng. Memseg: A semi- supervised method for image surface defect detection using differences and commonalities. Engineering Applications of Artificial Intelligence, 119:105835, 2023

    2023

  31. [39]

    Patch svdd: Patch-level svdd for anomaly detection and segmentation

    Jihun Yi and Sungroh Yoon. Patch svdd: Patch-level svdd for anomaly detection and segmentation. In Proceedings of the Asian conference on computer vision, 2020

    2020

  32. [40]

    Self-supervised learning for anomaly detection with dynamic local augmentation

    Seungdong Yoa, Seungjun Lee, Chiyoon Kim, and Hyun- woo J Kim. Self-supervised learning for anomaly detection with dynamic local augmentation. IEEE Access, 9:147201– 147211, 2021

    2021

  33. [41]

    Fastflow: Unsupervised anomaly detection and localization via 2d normalizing flows

    Jiawei Yu, Ye Zheng, Xiang Wang, Wei Li, Yushuang Wu, Rui Zhao, and Liwei Wu. Fastflow: Unsupervised anomaly detection and localization via 2d normalizing flows. arXiv preprint arXiv:2111.07677, 2021. 3

    2021 arXiv

  34. [42]

    Wide residual net- works

    Sergey Zagoruyko and Nikos Komodakis. Wide residual net- works. In BMVC, 2016. 6

    2016

  35. [43]

    Draem- a discriminatively trained reconstruction embedding for sur- face anomaly detection

    Vitjan Zavrtanik, Matej Kristan, and Danijel Skoˇcaj. Draem- a discriminatively trained reconstruction embedding for sur- face anomaly detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 8330– 8339, 2021. 3

    2021

  36. [44]

    Anomaly detection using improved deep svdd model with data structure preservation

    Zheng Zhang and Xiaogang Deng. Anomaly detection using improved deep svdd model with data structure preservation. Pattern Recognition Letters, 148:1–6, 2021. 3

    2021

  37. [45]

    Pad: A dataset and benchmark for pose-agnostic anomaly detection

    Qiang Zhou, Weize Li, Lihan Jiang, Guoliang Wang, Guyue Zhou, Shanghang Zhang, and Hao Zhao. Pad: A dataset and benchmark for pose-agnostic anomaly detection. arXiv preprint arXiv:2310.07716, 2023. 1, 3

    2023 arXiv

  38. [46]

    Spot-the-difference self-supervised pre- training for anomaly detection and segmentation

    Yang Zou, Jongheon Jeong, Latha Pemula, Dongqing Zhang, and Onkar Dabeer. Spot-the-difference self-supervised pre- training for anomaly detection and segmentation. arXiv preprint arXiv:2207.14315, 2022. 3 Supplemental material In this supplemental material, we provide addition...

    2022 arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.