Pith. sign in

REVIEW 3 major objections 5 minor 29 references

RoBiS: Robust Binary Segmentation for High-Resolution Industrial Images

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read RoBiS reports that a preprocessing-plus-thresholding pipeline lifts industrial defect segmentation F1 from 21.8% to 51.00% on the private MVTec AD 2 test set, and from 16.7% to 46.52% on the mixed-lighting private test set.

desk verdict A credible challenge entry that lifts SegF1 on MVTec AD 2 with known components, but the small-anomaly story is weaker than advertised. read the letter →

arxiv 2505.21152 v1 pith:33XX46XC submitted 2025-05-27 cs.CV

Xurui Li , Zhonesheng Jiang , Tingxuan Ai , Yu Zhou This is my paper · ORCID
classification cs.CV
keywords unsupervisedanomalydetectionindustrialdefectsegmentationhigh-resolutionimagepreprocessingsliding-windowcroppingadaptivebinarizationMVTecAD2lightingaugmentationSAMrefinement
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

This paper argues that the steep performance drop of unsupervised anomaly detection on the MVTec AD 2 industrial benchmark can be reversed by treating the full pipeline as three separate robustness problems: keeping small defects visible in high-resolution images, hardening the detector to lighting and noise shifts, and turning continuous anomaly maps into accurate binary masks. It reports that its RoBiS pipeline raises segmentation F1 on the private test set from 21.8% to 51.00% and on the mixed-lighting private test set from 16.7% to 46.52%, gains of 29.2 and 29.82 percentage points over the compared baselines. If the numbers hold, the practical message is that preprocessing and post-processing choices, not just the anomaly-detection model, are the main remaining bottleneck for real industrial deployment.

What carries the argument

The load-bearing mechanism is the combination of Swin-Cropping and the joint binarization cascade. Swin-Cropping is an overlapping sliding-window preprocessing step (1024x1024 windows, 10% overlap, zero-padded boundaries) that preserves small and elongated anomalies while avoiding boundary-split artifacts; it is what lets a detector trained on 518x518 sub-images see defects that occupy as few as five pixels in the original high-resolution frame. The binarization cascade first thresholds each merged anomaly map by the per-pixel mean plus three standard deviations, takes the logical OR with MEBin's adaptively thresholded mask, then refines the result with SAM prompted by bounding boxes of the coarse regions. This cascade is what converts continuous anomaly scores into the final binary masks that the SegF1 metric is computed on.

What would settle it

Measure SegF1 on MVTec AD 2 TESTpriv with Swin-Cropping disabled (full high-resolution images resized directly to 518x518) while keeping the same augmentation and binarization; if small-defect categories such as Can and Wall Plugs do not lose most of their already-low scores, the reported gains cannot be attributed to the cropping step. The official per-category scoreboard would reveal whether the 14.38% and 1.86% SegF1 on Wall Plugs and Can are actually driven by the surviving small defects.

Watch

Extended reading notes

Core claim

On the paper's own terms, RoBiS establishes that the central difficulty in high-resolution industrial anomaly segmentation is not the detector alone but the joint handling of scale, illumination, and thresholding. Swin-Cropping cuts each high-resolution image into overlapping 1024x1024 windows with 10% overlap, so that anomalies as small as a handful of pixels survive into the sub-images; those sub-images are resized to 518x518 and fed to INP-Former, a multi-class anomaly-detection baseline initialized with DINOv2-R features. During training, half the samples receive Gaussian noise and exposure shifts to simulate the lighting variations of the test sets. At inference, overlapping anomaly maps are merged, then binarized by the union of a mean-plus-three-standard-deviation threshold and the authors' MEBin adaptive threshold, and finally refined by prompting SAM with the bounding boxes of coarse abnormal regions. The paper reports SegF1 of 51.00% on TESTpriv and 46.52% on TESTpriv,mix, which it presents as a 29.2-percentage-point and a 29.82-percentage-point gain over the best compared methods.

Load-bearing premise

The pipeline assumes that the cropping-and-resizing step keeps defects as small as a few pixels detectable, meaning no tiny anomaly is destroyed or blurred below the detector's resolution before the model ever sees it.

Editorial extensions

If this is right

  • Overlapping window cropping alone yields a 3.7-percentage-point F1-max improvement on TESTpub by preserving small-anomaly information in sub-images.
  • Adding Gaussian noise and exposure-shift augmentation to 50% of training samples contributes another 1.9 percentage points on TESTpub.
  • MEBin's adaptive threshold adds 0.4 percentage points of SegF1 on TESTpub by catching subtle anomalies that the simple mean-plus-three-standard-deviations threshold misses.
  • SAM-Finer contributes 12.5 percentage points of SegF1 on TESTpub, mainly by filling internal false negatives in fabric-class defects.
  • The full pipeline reaches 51.00% SegF1 on TESTpriv and 46.52% on TESTpriv,mix, with balanced ClassF1 and AucPro0.05 scores, and the code is released.

Reading between the lines

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

  • A testable extension is to ablate the input resolution: because the pipeline resizes 1024x1024 crops down to 518x518, a detector that operates natively at 1024x1024 might recover even more of the small-anomaly signal, making the reported gains a lower bound on what preprocessing alone can do.
  • The same preprocessing-plus-adaptive-binarization recipe could transfer to other high-resolution anomaly benchmarks beyond MVTec AD 2, and the paper's segmentation gains suggest that thresholding is undervalued relative to detector architecture in the field.
  • Because SAM-Finer contributes the largest single post-processing gain (12.5 points on TESTpub), the paper implies that foundation-model refinement is a low-cost upgrade for any continuous anomaly map; attaching SAM-Finer to a different baseline's anomaly maps without retraining would test this directly.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The manuscript presents RoBiS, a pipeline for binary defect segmentation on the MVTec AD 2 benchmark. It combines three components: (1) Swin-Cropping, a sliding-window preprocessing strategy that crops 1024x1024 patches with 10% overlap from high-resolution images; (2) an INP-Former baseline trained with Gaussian noise and exposure-based photometric augmentation; and (3) a post-processing stage that merges sub-image anomaly maps, binarizes them with a combination of mean+3std and the authors' prior MEBin module, and refines the result with SAM. The authors report SegF1 improvements from 21.8% to 51.00% on Test_private and from 16.7% to 46.52% on Test_private_mixed, with all private-test metrics computed by the official evaluation platform. Ablations on the public TESTpub set are provided for the cropping, augmentation, MEBin, and SAM-Finer components.

Significance. If the reported private-test numbers are accurate, the results represent a substantial improvement over the published baselines on a challenging industrial benchmark, and the pipeline is simple enough to be useful in practice. The paper has concrete strengths: the headline SegF1 values are externally computed by the official challenge platform rather than self-reported, code is promised at a public repository, and the framework is decomposed into clearly described modules. However, the causal explanation for the improvement is not fully supported by the experiments as presented: the claim that Swin-Cropping preserves small-anomaly information is undermined by the subsequent resize step, and the ablation tables lack the no-module baseline needed to attribute gains to individual components. The paper would be strengthened by additional experiments that directly test the small-anomaly mechanism and by completing the ablation design.

major comments (3)
  1. [Sec. 2.1, Sec. 2.2.2, Table 1] The central design motivation for Swin-Cropping is that it 'preserves the information of small anomalies' by cropping high-resolution images, but the manuscript does not establish that this mechanism actually survives the subsequent resize. Section 2.1 reports that over 20% of anomalies occupy fewer than 281 pixels and that the smallest anomaly contains only 5 pixels; Section 2.2.2 then resizes every 1024x1024 crop to 518x518. A 5-pixel anomaly is reduced to roughly 1.3 square pixels after bilinear downsampling, and the authors themselves note in Section 3.2 that SAM's 64x64 feature resolution 'prevents more precise segmentation of tiny anomalies.' The category-level results in Table 1 are consistent with a failure on very small defects: Can (1.86/0.84) and Wall Plugs (14.38/3.40) are far below the mean, while texture categories such as Fabric and Rice dominate the average. To support the claimed mechanism, please provide a quantitative analysis of performance as a function of anomaly size, or an experiment comparing cropping with and without the 518x518 resize on categories with tiny defects, or a synthetic downsampling study that shows small anomalies are not destroyed before detection.
  2. [Tables 3 and 4] The ablation tables do not include a baseline row with neither module active, so the claimed contributions of individual components cannot be computed. In Table 3, the text states that Swin-Cropping gives a 3.7% F1-max improvement and data augmentation gives an additional 1.9%, but the table only shows three configurations (Swin only, augmentation only, and both); without a 'neither' row, the 3.7% is a marginal gain over augmentation alone and the 1.9% is a marginal gain over Swin-Cropping alone, not an isolation of either module. The same issue applies to Table 4, where the claimed 0.4% gain for MEBin and 12.5% gain for SAM-Finer require a no-module baseline that is absent. Additionally, these ablations are reported only on TESTpub, while the headline claims are on the private test sets; please add the missing baseline row and, where possible, report the corresponding ablations on the private sets or justify why TESTpub is representative.
  3. [Sec. 2.3] The choice of SAM variant is test-set dependent: the authors state 'we use SAM-Huge architecture for TESTpriv and SAM-Base for TESTpriv,mix' without giving a rationale or an ablation. This is a free design parameter applied differently to the two private test sets, and if the choice was made using feedback from the official evaluation platform, it affects the interpretation of the reported numbers as a fixed unsupervised pipeline. Please provide a justification for this choice, report public-set validation for both variants, or state explicitly how the variant was selected.
minor comments (5)
  1. [Sec. 3.2] The sentence beginning 'While our method successfully identifies both structural and logical anomalies, architectural constraints inherent to anomaly detection models, particularly the need for successive downsampling to achieve sufficient receptive fields.' is grammatically incomplete and should be finished.
  2. [Abstract and Sec. 5] The full text says 'Code is available at github link' instead of an actual URL; please replace this placeholder with the complete repository address.
  3. [Sec. 2.2.1] The term 'multi-class approaches' is confusing because the method trains one model per product category; please clarify that 'multi-class' refers to multiple sub-image appearances within a category rather than multiple semantic classes.
  4. [Sec. 2.4 and Table 1] The notation for the private test sets is inconsistent: the abstract uses Test_private and Test_private_mixed, while the body and tables use TESTpriv and TESTpriv,mix. Please unify the notation.
  5. [Reference [12]] The reference title 'The mvtec ad 2 dataset: Advanced scenarios for unsupervised anomaly detection' should be capitalized consistently as 'The MVTec AD 2 Dataset: Advanced Scenarios for Unsupervised Anomaly Detection'.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular derivation: central SegF1 claims are externally benchmarked; the only self-citation (MEBin) is ablated, not assumed into the result.

full rationale

The paper's headline results (51.00% SegF1 on TESTpriv, 46.52% on TESTpriv,mix) are computed by the official MVTec AD 2 challenge platform, an external benchmark, not by a formula defined in terms of RoBiS's own outputs. Each component is supported by ablations on TESTpub: Swin-Cropping and data augmentation in Table 3, and MEBin/SAM-Finer in Table 4. The single self-citation is MEBin (AnomalyNCD, CVPR 2025), used as one branch of adaptive binarization; Table 4 shows it adds only 0.4% SegF1 over mean+3std, while SAM-Finer contributes 12.5%, and the final numbers come from the full externally scored pipeline. This is a normal, non-round-tripping use of prior work: the target result is not defined in terms of MEBin, and the module is independently ablated rather than assumed. The Sec 3.2 admission that SAM's 64x64 feature resolution 'prevents more precise segmentation of tiny anomalies' and the Sec 2.1/2.2.2 concern that 1024x1024 crops are resized to 518x518 are correctness risks for the causal claim that Swin-Cropping preserves small anomalies, but they are not circularity: that claim is tested empirically and the SegF1 metric is external. No equation reduces to its own input, and no fitted parameter is renamed as a prediction.

Assumptions & free parameters 7 free parameters · 4 assumptions · 0 invented entities

The central claim depends on a set of hand-chosen hyperparameters and assumptions about the transferability of pretrained models and the correctness of the official evaluation. No new physical or conceptual entity is introduced.

free parameters (7)
  • Swin-Cropping window size = 1024x1024
    Chosen to balance efficiency and accuracy; no sensitivity analysis shown.
  • Swin-Cropping overlap = 10%
    Chosen to avoid boundary artifacts; no sensitivity analysis shown.
  • Sub-image resize resolution = 518x518
    All sub-images are resized to 518x518, which may discard small anomaly details.
  • Gaussian noise sigma = 15
    Noise augmentation strength chosen without tuning or justification.
  • Exposure lambda range = U(-0.2, 0.2)
    Exposure variation range hand-picked; no experiments on other ranges.
  • Augmentation fraction = 50%
    Applied to half of training samples; no justification given.
  • SAM variant per test set = Huge for TESTpriv, Base for TESTpriv,mix
    No stated criterion for the choice; could be selection on test performance.
assumptions (4)
  • domain assumption DINOv2-R pre-trained features transfer to industrial anomaly detection
    Used as encoder initialization in Sec 2.2.2.
  • domain assumption INP-Former is an appropriate base model for anomaly map generation
    Chosen as baseline without comparison to alternative backbones.
  • domain assumption Official MVTec AD 2 evaluation platform computes SegF1 correctly and consistently
    All private-set numbers come from the official platform, as stated in Sec 3.1.
  • domain assumption MEBin thresholding from prior work is effective on this benchmark
    Adaptive binarization relies on MEBin from the authors' earlier CVPR 2025 paper, with no independent validation here.

how reviews work

0 comments
Cite this review

Pith. "Pith review of RoBiS: Robust Binary Segmentation for High-Resolution Industrial Images." pith.science (2026). https://pith.science/paper/33XX46XC

@misc{pith2026250521152,
  author       = {Pith},
  title        = {Pith review of: RoBiS: Robust Binary Segmentation for High-Resolution Industrial Images},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/33XX46XC}},
  note         = {Machine review of arXiv:2505.21152}
}
read the original abstract

Robust unsupervised anomaly detection (AD) in real-world scenarios is an important task. Current methods exhibit severe performance degradation on the MVTec AD 2 benchmark due to its complex real-world challenges. To solve this problem, we propose a robust framework RoBiS, which consists of three core modules: (1) Swin-Cropping, a high-resolution image pre-processing strategy to preserve the information of small anomalies through overlapping window cropping. (2) The data augmentation of noise addition and lighting simulation is carried out on the training data to improve the robustness of AD model. We use INP-Former as our baseline, which could generate better results on the various sub-images. (3) The traditional statistical-based binarization strategy (mean+3std) is combined with our previous work, MEBin (published in CVPR2025), for joint adaptive binarization. Then, SAM is further employed to refine the segmentation results. Compared with some methods reported by the MVTec AD 2, our RoBiS achieves a 29.2% SegF1 improvement (from 21.8% to 51.00%) on Test_private and 29.82% SegF1 gains (from 16.7% to 46.52%) on Test_private_mixed. Code is available at https://github.com/xrli-U/RoBiS.

Figures

Figures reproduced from arXiv: 2505.21152 by the authors.

Figure 1
Figure 1. The pipeline of our RoBiS. This framework contains three important parts: (1) Swin-Cropping for dataset pre-processing (Sec. 2.1). (2) INP-Former [17] to detect anomalies (Sec. 2.2). (3) Anomaly Maps Merging to obtain the anomaly map of original image, and Adaptive Binarization to generate its corresponding binary mask (Sec. 2.3). it is necessary to study the adaptive binarization method. Some researches [23, 25] in… view at source ↗
Figure 2
Figure 2. Visualization of anomaly segmentation results on the [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 27 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 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024. 3, 4

    work page 2024

  2. [2]

    Correcting deviations from normality: A reformulated diffusion model for multi-class unsupervised anomaly detection

    Farzad Beizaee, Gregory A Lodygensky, Christian Desrosiers, and Jose Dolz. Correcting deviations from normality: A reformulated diffusion model for multi-class unsupervised anomaly detection. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025. 2

    work page 2025

  3. [3]

    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 IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition (CVPR),

    work page

  4. [4]

    A survey on visual anomaly detection: Challenge, approach, and prospect

    Yunkang Cao, Xiaohao Xu, Jiangning Zhang, Yuqi Cheng, Xiaonan Huang, Guansong Pang, and Weiming Shen. A survey on visual anomaly detection: Challenge, approach, and prospect. In arXiv preprint arXiv:2401.16402, 2024. 1

    arXiv 2024

  5. [5]

    A unified anomaly synthesis strategy with gradient ascent for industrial anomaly detection and localization

    Qiyu Chen, Huiyuan Luo, Chengkan Lv, and Zhengtao Zhang. A unified anomaly synthesis strategy with gradient ascent for industrial anomaly detection and localization. In Euro- pean Conference on Computer Vision (ECCV), pages 37–54. Springer, 2024. 2

    work page 2024

  6. [6]

    Vision transformers need registers

    Timoth´ee Darcet, Maxime Oquab, Julien Mairal, and Piotr Bojanowski. Vision transformers need registers. In Interna- tional Conference on Learning Representations (ICLR), 2024. 2

    work page 2024

  7. [7]

    Anomaly detection via reverse distillation from one-class embedding

    Hanqiu Deng and Xingyu Li. Anomaly detection via reverse distillation from one-class embedding. In IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition (CVPR), pages 9737–9746, 2022. 3, 4

    work page 2022

  8. [8]

    Transfusion–a transparency-based diffusion model for anomaly detection

    Matic Fu ˇcka, Vitjan Zavrtanik, and Danijel Sko ˇcaj. Transfusion–a transparency-based diffusion model for anomaly detection. In European Conference on Computer Vision (ECCV), pages 91–108. Springer, 2024. 2

    work page 2024

Show all 29 references
  1. [9]

    Encoder-decoder contrast for unsupervised anomaly detection in medical images

    Jia Guo, Shuai Lu, Lize Jia, Weihang Zhang, and Huiqi Li. Encoder-decoder contrast for unsupervised anomaly detection in medical images. IEEE Transactions on Medical Imaging,

  2. [10]

    Dinomaly: The less is more philosophy in multi-class unsupervised anomaly detection

    Jia Guo, Shuai Lu, Weihang Zhang, Fang Chen, Huiqi Li, and Hongen Liao. Dinomaly: The less is more philosophy in multi-class unsupervised anomaly detection. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025. 2

    2025

  3. [11]

    Mambaad: Exploring state space models for multi-class unsupervised anomaly detec- tion

    Haoyang He, Yuhu Bai, Jiangning Zhang, Qingdong He, Hongxu Chen, Zhenye Gan, Chengjie Wang, Xiangtai Li, Guanzhong Tian, and Lei Xie. Mambaad: Exploring state space models for multi-class unsupervised anomaly detec- tion. Advances in Neural Information Processing Systems (Neur...

    2024

  4. [12]

    The mvtec ad 2 dataset: Ad- vanced scenarios for unsupervised anomaly detection

    Lars Heckler-Kram, Jan-Hendrik Neudeck, Ulla Scheler, Re- becca K¨onig, and Carsten Steger. The mvtec ad 2 dataset: Ad- vanced scenarios for unsupervised anomaly detection. arXiv preprint arXiv:2503.21622, 2025. 1, 2, 3

    2025

  5. [13]

    Anomalyncd: Towards novel anomaly class discovery in industrial scenarios

    Ziming Huang, Xurui Li, Haotian Liu, Feng Xue, Yuzhe Wang, and Yu Zhou. Anomalyncd: Towards novel anomaly class discovery in industrial scenarios. In IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition (CVPR),

  6. [14]

    Segment any- thing

    Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer White- head, Alexander C Berg, Wan-Yen Lo, et al. Segment any- thing. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 4015–4026, 2023. 3

    2023

  7. [15]

    Reducing boundary artifacts in image deconvolution

    Renting Liu and Jiaya Jia. Reducing boundary artifacts in image deconvolution. In IEEE International Conference on Image Processing (ICIP), pages 505–508. IEEE, 2008. 2

    2008

  8. [16]

    Simplenet: A simple network for image anomaly detection and localization

    Zhikang Liu, Yiming Zhou, Yuansheng Xu, and Zilei Wang. Simplenet: A simple network for image anomaly detection and localization. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 20402–20411,

  9. [17]

    Explor- ing intrinsic normal prototypes within a single image for universal anomaly detection

    Wei Luo, Yunkang Cao, Haiming Yao, Xiaotian Zhang, Jianan Lou, Yuqi Cheng, Weiming Shen, and Wenyong Yu. Explor- ing intrinsic normal prototypes within a single image for universal anomaly detection. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025. 2

    2025

  10. [18]

    Abnormal be- havior recognition for intelligent video surveillance systems: A review

    Amira Ben Mabrouk and Ezzeddine Zagrouba. Abnormal be- havior recognition for intelligent video surveillance systems: A review. Expert Systems with Applications , 91:480–491,

  11. [19]

    A threshold selection method from gray- level histograms

    Nobuyuki Otsu. A threshold selection method from gray- level histograms. IEEE Transactions on Systems, Man, and Cybernetics (TSMC), 9(1):62–66, 1979. 2

    1979

  12. [20]

    Towards total recall in industrial anomaly detection

    Karsten Roth, Latha Pemula, Joaquin Zepeda, Bernhard Sch¨olkopf, Thomas Brox, and Peter Gehler. Towards total recall in industrial anomaly detection. In IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition (CVPR),

  13. [21]

    Revisiting reverse distillation for anomaly detection

    Tran Dinh Tien, Anh Tuan Nguyen, Nguyen Hoang Tran, Ta Duc Huy, Soan Duong, Chanh D Tr Nguyen, and Steven QH Truong. Revisiting reverse distillation for anomaly detection. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023. 3, 4

    2023

  14. [22]

    Stable and low- precision training for large-scale vision-language models

    Mitchell Wortsman, Tim Dettmers, Luke Zettlemoyer, Ari Morcos, Ali Farhadi, and Ludwig Schmidt. Stable and low- precision training for large-scale vision-language models. Ad- vances in Neural Information Processing Systems (NeurIPS), 36:10271–10298, 2023. 3

    2023

  15. [23]

    Diffumask: Synthesizing images with pixel-level annotations for semantic segmentation using diffu- sion models

    Weijia Wu, Yuzhong Zhao, Mike Zheng Shou, Hong Zhou, and Chunhua Shen. Diffumask: Synthesizing images with pixel-level annotations for semantic segmentation using diffu- sion models. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 1206–1217, 2023. 2

    2023

  16. [24]

    Im-iad: Indus- trial image anomaly detection benchmark in manufacturing

    Guoyang Xie, Jinbao Wang, Jiaqi Liu, Jiayi Lyu, Yong Liu, Chengjie Wang, Feng Zheng, and Yaochu Jin. Im-iad: Indus- trial image anomaly detection benchmark in manufacturing. IEEE Transactions on Cybernetics, pages 1–14, 2024. 1

    2024

  17. [25]

    Dual- level adaptive self-labeling for novel class discovery in point cloud segmentation

    Ruijie Xu, Chuyu Zhang, Hui Ren, and Xuming He. Dual- level adaptive self-labeling for novel class discovery in point cloud segmentation. In European Conference on Computer Vision, pages 288–305. Springer, 2024. 2

    2024

  18. [26]

    Glad: towards better reconstruction with 6 global and local adaptive diffusion models for unsupervised anomaly detection

    Hang Yao, Ming Liu, Zhicun Yin, Zifei Yan, Xiaopeng Hong, and Wangmeng Zuo. Glad: towards better reconstruction with 6 global and local adaptive diffusion models for unsupervised anomaly detection. In European Conference on Computer Vision (ECCV), pages 1–17. Springer, 2024. 2

    2024

  19. [27]

    Dsr–a dual subspace re-projection network for surface anomaly de- tection

    Vitjan Zavrtanik, Matej Kristan, and Danijel Skoˇcaj. Dsr–a dual subspace re-projection network for surface anomaly de- tection. In European Conference on Computer Vision (ECCV),

  20. [28]

    Realnet: A feature selection network with realistic synthetic anomaly for anomaly detection

    Ximiao Zhang, Min Xu, and Xiuzhuang Zhou. Realnet: A feature selection network with realistic synthetic anomaly for anomaly detection. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 16699–16708,

  21. [29]

    Msflow: Multiscale flow-based framework for unsupervised anomaly detection

    Yixuan Zhou, Xing Xu, Jingkuan Song, Fumin Shen, and Heng Tao Shen. Msflow: Multiscale flow-based framework for unsupervised anomaly detection. IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2024. 3, 4 7

    2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.