浏览全部资源
扫码关注微信
网络出版日期： 2025-02-17 ，
录用日期： 2025-02-13
移动端阅览
刘建明,庄维宽.结合视觉Mamba和块特征分布的工业异常检测[J].中国图象图形学报,
Liu Jianming,Zhuang Weikuan.Industrial anomaly detection by combining visual mamba and patch feature distribution[J].Journal of Image and Graphics,
目的
2
工业异常检测在现代工业生产中具有至关重要的作用，现有的工业异常检测方法主要是基于卷积神经网络（convolutional neural network， CNN） 或视觉变换器（vision transformer， ViT）网络来实现。然而，CNN 存在难以处理长距离依赖关系的不足，而ViT又面临时间复杂度高的问题。基于此，提出了一种结合视觉Mamba和块特征分布的无监督工业异常检测模型。
方法
2
该模型包含两个互补分支网络：块特征分布估计网络和基于视觉Mamba的自编码重建网络。块特征分布估计网络主要依赖局部块特征来进行异常检测，通过融合高效的预训练块特征描述网络以及视觉Mamba编码器提取的正常样本的块特征，学习一个高斯混合密度网络来估计正常样本局部块特征的分布。在测试阶段利用这个高斯混合密度网络估计异常图像的各个位置的异常得分，从而得到一个局部异常得分图（（local anomaly map， LAM）；基于视觉Mamba的自编码重构网络则利用视觉Mamba编码器来捕捉长距离关联特征，增强跨不同类别和形态的复杂异常图像的全局建模能力，在测试阶段利用重建误差来估计异常图像的全局异常得分图（global anomaly map， GAM）；最后，合并LAM和GAM得到最终检测结果。
结果
2
在MvTec AD、VisA和BTAD等公开数据集上与其它最先进的算法进行了比较，取得了有竞争力的效果。在MvTec AD数据集上所提模型相比性能第二的模型在像素级 AU-ROC（Area Under the Receiver Operating Characteristic Curve， AU-ROC）指标提升了0.9%，在图像级别上 AU-ROC指标提升了2.4%。在BTAD数据集所提模型相比性能第二的模型在图像级上 AU-ROC 提升 0.4%。在VisA数据集上模型相比性能第二的模型在像素级 AU-ROC指标提升了0.6%。
结论
2
将视觉状态空间用于图像重建检测图像异常是可行的，检测效果具有竞争力。
Objective
2
Industrial image anomaly detection plays a crucial role in modern industrial production， as it can timely detect defects in products， effectively improve product qualification rate， enhance industrial productivity， and reduce production costs. Traditional anomaly detection algorithms often show certain limitations when facing new types of anomalies， especially complex issues such as logical anomalies， making it difficult to meet the demand for high-precision and efficient detection in industrial production. Therefore， this study is committed to exploring the potential application of visual state space in the field of image processing and anomaly detection， aiming to find a more effective method to address the shortcomings of traditional algorithms in detecting new types of anomalies， especially the limitations in dealing with logical anomalies. The reconstruction-based method is considered capable of addressing logical anomalies caused by factors such as object quantity， structure， position， and arrangement order because using only normal images to train the model will result in significant errors in the reconstructed output compared to images with logical anomalies. Existing reconstruction-based anomaly detection methods are mainly based on convolutional neural networks （CNN） or vision transformer （ViT） networks. However， CNN has the disadvantage of being difficult to handle long-distance dependencies， while ViT faces the problem of high time complexity. The latest research shows that state space models（SSM） represented by Mamba can effectively model long dependencies while maintaining linear complexity. We have explored the potential application of visual state space in anomaly detection and hope to develop a more precise and efficient image anomaly detection technology by leveraging its advantages to meet strict quality control requirements in industrial production. This will drive industrial production towards intelligent automation direction while improving overall efficiency and competitiveness.
Method
2
A novel unsupervised industrial anomaly detection model combining visual Mamba and patch feature distribution is proposed. This model consists of two complementary branch networks： a patch feature distribution estimation network and a self-encoding reconstruction network based on visual Mamba. The patch feature distribution estimation network primarily relies on local patch features for anomaly detection， fusing local patch features of normal samples through the Vision Mamba encoder and pretrained efficient patch description network and learning a Gaussian mixture density network to estimate the distribution of these features. During the testing phase， this Gaussian mixture density network is used to estimate anomaly scores at various positions in the anomalous images， resulting in a local anomaly map （LAM）. Meanwhile， the self-encoding reconstruction network based on visual Mamba utilizes a visual Mamba encoder to capture long-range associated features， enhancing the global modeling capability for complex anomaly detection across different categories and forms. In the testing phase， reconstruction errors are used to estimate a global anomaly map （GAM） for the anomalous images. Finally， LAM and GAM are combined to obtain the final detection results. For the data set， we carried out detailed preprocessing and clipped the images to appropriate sizes according to the requirements of different models. For example， the size of the input image was 256*256. We carefully adjust the number of coding blocks in the encoder of the visual state space in the reconstruction method to achieve the best anomaly detection performance and maximize the overall performance of the model. The experiments in this study were conducted on a desktop computer equipped with an Intel Core i5， 2.5GHz CPU， GeForce GTX 3060Ti GPU with 12GB memory， 32GB RAM， and running the Ubuntu18.04 operating system. According to our experimental observations， we have set the learning rate to 0.001， configured the model to run for 200 epochs， and determined a batch size of 48. Regarding the selection of image blocks， in the PDN method combined with Patchsize， we have chosen a value of 32.
Result
2
We compared our model with other advanced algorithms on publicly available datasets such as MvTec AD， VisA， and BTAD， and our model demonstrated highly competitive performance. On the MvTec AD dataset， our model improved the Pixel-level AU-ROC metric by 0.9% to reach 93.9%， and the Image-level AU-ROC metric by 2.4% to reach 93.8%， compared to the second-best performing model. On the BTAD dataset， our model achieved a 0.4% improvement in Image-level AU-ROC compared to the second-best performing model， reaching 92.6%. On the VisA dataset， our model achieved a 0.6% improvement in Pixel-level AU-ROC compared to the second-best performing model， reaching 96.6%. According to visualizations of anomaly localization in our paper on both MvTec and VisA datasets， our model's anomaly localization is more accurate than other models.
Conclusion
2
The application of visual state space to image reconstruction for detecting image anomalies is a feasible and effective method， and its anomaly localization effect has significant competitiveness. In addition， this study believes that aggregating features in the middle of the extraction model will be more helpful for adapting to anomaly detection tasks， and we believe that the setting of the number of image block vectors may be helpful for the localization and detection of anomalies because more image block descriptor vectors can represent more detailed information. These two points are worth further research in the future. This paper organically combines two popular methods in the industrial anomaly detection field， while integrating visual state space into the model， exploring its application in the field of anomaly detection.
Bae J ， Lee J - H and Kim S . 2023 . Pni： industrial anomaly detection using position and neighborhood information // Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision （ICCV） . Paris， France ： IEEE： 6350 - 6360 ［ DOI： 10.1109/ICCV51070.2023.00586 http://dx.doi.org/10.1109/ICCV51070.2023.00586 ］
Batzner K ， Heckler L and Konig R . 2024 . Efficient ad： accurate visual anomaly detection at millisecond - level latencies // Proceedings of the 2024 IEEE/CVF Winter Conference on Applications of Computer Vision （WACV） . Los Alamitos， CA， USA ： IEEE Computer Society： 127 - 137 ［ DOI： 10.1109/WACV57701.2024.00020 http://dx.doi.org/10.1109/WACV57701.2024.00020 ］
Bergmann P ， Batzner K ， Fauser M ， Sattlegger D and Steger C . 2022 . Beyond dents and scratches： logical constraints in unsupervised anomaly detection and localization . International Journal of Computer Vision ， 130 （ 4 ）： 947 - 969 ［ DOI： 10.1007/s11263 - 022 - 01578 - 9 http://dx.doi.org/10.1007/s11263-022-01578-9 ］
Bergmann P ， Löwe S ， Fauser M ， Sattlegger D and Steger C . 2019 . Improving unsupervised defect segmentation by applying structural similarity to autoencoders // Proceedings of the 14th International Joint Conference on Computer Vision， Imaging and Computer Graphics Theory and Applications （VISIGRAPP 2019） - Volume 5： VISAPP . SciTePress ： 372 - 380 ［ DOI： 10.5220/0007364503720380 http://dx.doi.org/10.5220/0007364503720380 ］
Bergmann P ， Fauser M ， Sattlegger D and Steger C . 2020 . Uninformed students： student - teacher anomaly detection with discriminative latent embeddings // Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR） . Los Alamitos， CA， USA ： IEEE Computer Society： 4182 - 4191 ［ DOI： 10.1109/CVPR42600.2020.00424 http://dx.doi.org/10.1109/CVPR42600.2020.00424 ］
Bergmann P ， Fauser M ， Sattlegger D and Steger C . 2019 . Mvtec ad–a comprehensive real-world dataset for unsupervised anomaly detection // Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR） . Los Alamitos， CA， USA ： IEEE Computer Society： 9584 - 9592 ［ DOI： 10.1109/CVPR.2019.00982 http://dx.doi.org/10.1109/CVPR.2019.00982 ］
Bishop C M . 1994 . Mixture density networks . ISBN ： NCRG/ 94 /004. Aston University， Birmingham
Cohen N and Hoshen Y . 2020 . Sub - image anomaly detection with deep pyramid correspondences ［EB/OL］.［ 2021-2-3 ］. https://doi.org/10.48550/arXiv.2005.02357.pdf https://doi.org/10.48550/arXiv.2005.02357.pdf
Deng H and Li X . 2022 . anomaly detection via reverse distillation from one-class embedding // Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR） . New Orleans， LA， USA ： IEEE： 9727 - 9736 ［ DOI： 10.1109/CVPR52688.2022.00951 http://dx.doi.org/10.1109/CVPR52688.2022.00951 ］
Defard T ， Setkov A ， Loesch A and Audigier R . 2021 . Padim： a patch distribution modeling framework for anomaly detection and localization // Proceedings of the International Conference on Pattern Recognition （ICPR） . Milano， Italy ： Springer： 475 - 489 ［ DOI： 10.1007/978-3-030-68799-1_35 http://dx.doi.org/10.1007/978-3-030-68799-1_35 ］
Fang Z ， Wang X Y ， Li H C ， Liu J J ， Hu Q G and Xiao J M . 2023 . Fastrecon： few - shot industrial anomaly detection via fast feature reconstruction // Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision （ICCV） . Paris， France ： IEEE： 17435 - 17444 ［ DOI： 10.1109/ICCV51070.2023.01603 http://dx.doi.org/10.1109/ICCV51070.2023.01603 ］
Gu A ， Goel K and Ré C . 2022 . Efficiently modeling long sequences with structured state spaces ［EB/OL］.［ 2022-8-5 ］. https://doi.org/10.48550/arXiv.2111.00396.pdf https://doi.org/10.48550/arXiv.2111.00396.pdf
Gu A ， Dao T . 2023 . Mamba： linear - time sequence modeling with selective state spaces ［EB/OL］. ［ 2024-5-31 ］. https://doi.org/10.48550/arXiv.2312.00752.pdf https://doi.org/10.48550/arXiv.2312.00752.pdf
Gudovskiy D ， Ishizaka S and Kozuka K . 2022 . Cflow - ad： real - time unsupervised anomaly detection with localization via conditional normalizing flows // Proceedings of the 2022 IEEE/CVF Winter Conference on Applications of Computer Vision （WACV） . Waikoloa， HI， USA ： IEEE： 1819 - 1828 ［ DOI： 10.1109/WACV51458.2022.00188 http://dx.doi.org/10.1109/WACV51458.2022.00188 ］
Gu Z H ， Liu L ， Chen X ， Yi R ， Zhang J N ， Wang Y B ， Wang C G ， Shu A N ， Jiang G and Ma L Z . 2023 . Remembering normality： memory-guided knowledge distillation for unsupervised anomaly detection // Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision （ICCV） . Paris， France ： IEEE： 16355 - 16363 ［ DOI： 10.1109/ICCV51070.2023.01503 http://dx.doi.org/10.1109/ICCV51070.2023.01503 ］
Huang C ， Guan H ， Jiang A ， Zhang Y ， Spratling M and Wang Y F . 2022 . Registration based few - shot anomaly detection // European Conference on Computer Vision （ECCV 2022） . Switzerland ： Springer Cham： 303 - 319 ［ DOI： 10.1007/978 - 3 - 031 - 20053 - 3_18 http://dx.doi.org/10.1007/978-3-031-20053-3_18 ］
Jeong J ， Zou Y ， Kim T ， Zhang D ， Ravichandran A and Dabeer O . 2023 . Winclip： zero-/few-shot anomaly classification and segmentation // Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR） . Los Alamitos， CA， USA ： IEEE Computer Society： 19606 - 19616 ［ DOI： 10.1109/CVPR52729.2023.01878 http://dx.doi.org/10.1109/CVPR52729.2023.01878 ］
Kim D ， Park C ， Cho S and Lee S . 2023 . Fapm： fast adaptive patch memory for real - time industrial anomaly detection // Proceedings of the 2023 IEEE International Conference on Acoustics， Speech and Signal Processing （ICASSP） . Rhodes Island， Greece ： IEEE： 1 - 5 ［ DOI： 10.1109/ICASSP49357.2023.10096400 http://dx.doi.org/10.1109/ICASSP49357.2023.10096400 ］
Lee Y and Kang P . 2022 . Anovit： unsupervised anomaly detection and localization with vision transformer-based encoder-decoder. IEEE Access . IEEE ： 10 ： 46717 - 46724 ［ DOI： 10.1109/ACCESS.2022.3171559 http://dx.doi.org/10.1109/ACCESS.2022.3171559 ］
Liu Z ， Zhou Y ， Xu Y and Wang Z . 2023 . Simplenet： a simple network for image anomaly detection and localization // Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR） . Los Alamitos， CA， USA ： IEEE Computer Society： 20402 - 20411 ［ DOI： 10.1109/CVPR52729.2023.01954 http://dx.doi.org/10.1109/CVPR52729.2023.01954 ］
Mathian E ， Liu H ， Fernandez-Cuesta L ， Samaras D ， Foll M and Chen L . 2023 . Haloae： a local transformer auto - encoder for anomaly detection and localization based on halonet // Proceedings of the 18th International Joint Conference on Computer Vision， Imaging and Computer Graphics Theory and Applications （VISIGRAPP 2023） . Lisbon， Portugal ： SCITEPRESS： 325 - 337 ［ DOI： 10.5220/0011865900003417 http://dx.doi.org/10.5220/0011865900003417 ］
Mishra P ， Verk R ， Fornasier D ， Piciarelli C and Foresti G L . 2021 . Vt-adl： a vision transformer network for image anomaly detection and localization // Proceedings of the 2021 IEEE 30th International Symposium on Industrial Electronics （ISIE） . Kyoto， Japan ： IEEE： 01 - 06 ［ DOI： 10.1109/isie45552.2021.9576231 http://dx.doi.org/10.1109/isie45552.2021.9576231 ］
McIntosh D and Albu A B . 2023 . Inter-realization channels： unsupervised anomaly detection beyond one-class classification // Proceedings of the IEEE/CVF International Conference on Computer Vision （ICCV） . Paris， France ： IEEE： 6262 - 6272 . ［ DOI： 10.1109/ICCV51070.2023.00578 http://dx.doi.org/10.1109/ICCV51070.2023.00578 ］
Pang G ， Shen C ， Cao L and Hengel A V D . 2022 . Deep learning for anomaly detection： a review . ACM Computing Surveys ， 54 （ 2 ）： 38 - 75 ［ DOI： 10.1145/3439950 http://dx.doi.org/10.1145/3439950 ］
P. Mishra ， R. Verk ， D. Fornasier ， C. Piciarelli ， and G.L. Foresti（ 2021 ）. VT-ADL： A Vision Transformer Network for Image Anomaly Detection and Localization ［C］，30th IEEE/IES International Symposium on Industrial Electronics （ISIE），Kyoto， Japan， June 20 - 23 .［ DOI： 10.1109/ISIE45552.2021.9576231 http://dx.doi.org/10.1109/ISIE45552.2021.9576231 ］
Pirnay J and Chai K . 2022 . Inpainting transformer for anomaly detection // Proceedings of the International Conference on Image Analysis and Processing . Berlin， Heidelberg ： Springer - Verlag： 394 - 406 ［ DOI： 10.1007/978-3-031-06430-2_33 http://dx.doi.org/10.1007/978-3-031-06430-2_33 ］
Rudolph M ， Wandt B and Rosenhahn B . 2021 . Same same but differnet： semi-supervised defect detection with normalizing flows // Proceedings of the 2021 IEEE Winter Conference on Applications of Computer Vision （WACV） . Waikoloa， HI， USA ： IEEE Computer Society： 1906 - 1915 ［ DOI： 10.1109/WACV48630.2021.00195 http://dx.doi.org/10.1109/WACV48630.2021.00195 ］
Rudolph M ， Wehrbein T ， Rosenhahn B and Wandt B . 2023 . Asymmetric Student-Teacher Networks for Industrial Anomaly Detection // Proceedings of the 2023 IEEE/CVF Winter Conference on Applications of Computer Vision （WACV） . Waikoloa， HI， USA ： IEEE Computer Society： 2591 - 2601 ［ DOI： 10.1109/WACV56688.2023.00262 http://dx.doi.org/10.1109/WACV56688.2023.00262 ］
Rippel O ， Mertens P and Merhof D . 2021 . Modeling the distribution of normal data in pre-trained deep features for anomaly detection // Proceedings of the 2020 25th International Conference on Pattern Recognition （ICPR） . Milan， Italy ： IEEE Computer Society： 6726 - 6733 ［ DOI： 10.1109/ICPR48806.2021.9412109 http://dx.doi.org/10.1109/ICPR48806.2021.9412109 ］
Roth K ， Pemula L ， Zepeda J ， Scholkopf B ， Brox T and Gehler P . 2022 . Towards total recall in industrial anomaly detection // Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR） . New Orleans， LA， USA ： IEEE Computer Society： 14298 - 14308 ［ DOI： 10.1109/CVPR52688.2022.01392 http://dx.doi.org/10.1109/CVPR52688.2022.01392 ］
Sheynin S ， Benaim S and Wolf L . 2021 . A hierarchical transformation-discriminating generative model for few shot anomaly detection // Proceedings of the IEEE/CVF International Conference on Computer Vision （ICCV） . Montreal， QC， Canada ： IEEE： 8495 - 8504 ［ DOI： 10.1109/ICCV48922.2021.00838 http://dx.doi.org/10.1109/ICCV48922.2021.00838 ］
Smith J T H ， Warrington A and Linderman S W . 2023 . Simplified state space layers for sequence modeling ［EB/OL］. ［ 2023-3-3 ］. https://doi.org/10.48550/arXiv.2208.04933.pdf https://doi.org/10.48550/arXiv.2208.04933.pdf
Sugawara S and Imamura R . 2024 . Puad： frustratingly simple method for robust anomaly detection // Proceedings of the 2024 IEEE International Conference on Image Processing （ICIP） . Abu Dhabi， United Arab Emirates ： IEEE： 842 - 848 ［ DOI： 10.1109/ICIP51287.2024.10647438 http://dx.doi.org/10.1109/ICIP51287.2024.10647438 ］
Wang S Q ， Cheng C ， Shi M and Zhu D M . 2024 . Defect detection method for industrial product surfaces with similar features by combining frequency and vit . Journal of Image and Graphics ， 29 （ 10 ）： 3074 - 3089
王素琴 ， 程成 ， 石敏 ， 朱登明 . 2024 . 结合频率和ViT的工业产品表面相似特征缺陷检测方法 . 中国图象图形学报 ， 29 （ 10 ）： 3074 - 3089 ［ DOI： 10.11834/jig.230532 http://dx.doi.org/10.11834/jig.230532 ］
Yang Q and Guo R . 2024 . An unsupervised method for industrial image anomaly detection with vision transformer-based autoencoder . Sensors ， 24 （ 8 ）： 2440 ［ DOI： 10.3390/s24082440 http://dx.doi.org/10.3390/s24082440 ］
Yang Zou ， Jongheon Jeong ， Latha Pemula ， Dongqing Zhang ， and Onkar Dabeer（ 2022 ）. Spot-the-difference self-supervised pretraining for anomaly detection and segmentation . In European Conference on Computer Vision， pages 392 – 408 .［ DOI： 10.1007/978-3-031-20056-4_23 http://dx.doi.org/10.1007/978-3-031-20056-4_23 ］
Yi J and Yoon S . 2021 . Patch svdd： patch-level svdd for anomaly detection and segmentation ［c］// Proceedings of the Asian Conference on Computer Vision （ACCV） 2020 . Lecture Notes in Computer Science， vol 12627 . Switzerland ： Springer Cham： 375 - 390 ［ DOI： 10.1007/978-3-030-69544-6_23 http://dx.doi.org/10.1007/978-3-030-69544-6_23 ］
Yu J ， Zheng Y ， Wang X ， Li W ， Wu Y ， Zhao R and Wu L . 2021 . Fastflow： unsupervised anomaly detection and localization via 2d normalizing flows ［EB/OL］.［ 2021-11-16 ］. https://arxiv.org/abs/2111.07677 https://arxiv.org/abs/2111.07677
Zavrtanik V ， Kristan M and Skočaj D . 2022 . Dsr – a dual subspace re-projection network for surface anomaly detection // Proceedings of the 17th European Conference on Computer Vision （ECCV） . Switzerland ： Springer Cham： 539 – 554 ［ DOI： 10.1007/978-3-031-19821-2_31 http://dx.doi.org/10.1007/978-3-031-19821-2_31 ］
Zavrtanik V ， Kristan M and Skočaj D . 2021 . Draem： a discriminatively trained reconstruction embedding for surface anomaly detection // Proceedings of the IEEE/CVF International Conference on Computer Vision （ICCV） . Montreal， QC，Canada ： IEEE： 8310 - 8319 ［ DOI： 10.1109/ICCV48922.2021.00822 http://dx.doi.org/10.1109/ICCV48922.2021.00822 ］
Zhang X ， Li S ， Li X ， Huang P ， Shan J and Chen T . 2023 . Destseg： segmentation guided denoising student - teacher for anomaly detection // Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR） . Vancouver， BC， Canada ： IEEE： 3914 - 3923 ［ DOI： 10.1109/CVPR52729.2023.00381 http://dx.doi.org/10.1109/CVPR52729.2023.00381 ］
Zhu L ， Liao B ， Zhang Q ， Wang X ， Liu W Y and Wang X G . 2025 . Vision mamba： efficient visual representation learning with bidirectional state space model // Proceedings of the 41st International Conference on Machine Learning （ICML） . Vienna， Austria ： JMLR.org： 62429 - 62442 ［ DOI： 10.5555/3692070.3694654 http://dx.doi.org/10.5555/3692070.3694654 ］
Zhao Y . 2023 . Omnial： a unified cnn framework for unsupervised anomaly localization // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR） . Vancouver， BC， Canada ： IEEE： 3924 - 3933 . ［ DOI： 10.1109/CVPR52729.2023.00382 http://dx.doi.org/10.1109/CVPR52729.2023.00382 ］
Zhang X ， Li N ， Li J ， Dai T ， Jiang Y and Xia S T . 2023 . Unsupervised surface anomaly detection with diffusion probabilistic model // Proceedings of the IEEE/CVF International Conference on Computer Vision （ICCV） . Paris， France ： IEEE： 6759 - 6768 ［ DOI： 10.1109/ICCV51070.2023.00624 http://dx.doi.org/10.1109/ICCV51070.2023.00624 ］
Zhou K ， Xiao Y T ， Yang J L ， Cheng J ， Liu W ， Luo W X ， Gu Z W ， Liu J and Gao S H . 2020 . Encoding structure-texture relation with p-net for anomaly detection in retinal images // Proceedings of the European Conference on Computer Vision （ECCV） . Switzerland ： Springer： 360 - 377 ［ DOI： 10.1007/978-3-030-58565-5_22 http://dx.doi.org/10.1007/978-3-030-58565-5_22 ］
Zou Y ， Jeong J ， Pemula L ， Zhang D and Dabeer O . 2022 . Spot-the-difference self-supervised pre-training for anomaly detection and segmentation // Proceedings of the European Conference on Computer Vision （ECCV） . Switzerland ： Springer Cham： 392 - 408 ［ DOI： 10.1007/978-3-031-20056-4_23 http://dx.doi.org/10.1007/978-3-031-20056-4_23 ］
相关作者
相关机构
京公网安备11010802024621