在異常檢測中，希望透過電腦來分辨出異常的物體和正常的物體，來實現自動品質管理的功能，但是由於異常樣本取得不易，不光是異常的樣本難以蒐集，另外還有其異常的類別難以全部收集，因此相關的異常檢測模型傾向於以正常樣本來訓練模型，也就是透過無監督學習的方式，去學習並得到一個專屬模型，而其模型可以用於預測測試試資料是否為異常樣本。然而在以往的異常檢測模型，通常都是使用神經網路，用於訓練的正常樣本需要好幾張的圖片，以避免過擬合。然而要收集這麼多的圖片才能進行訓練。往往人眼只需透過幾張正常圖片就可以進行識別，因此本篇論文欲使用少量樣本以進行異常偵測，也就是少樣本異常檢測，此領域出現的原因。
本篇論文以標準化流為靈感，去實現少樣本的異常偵測，透過機率分佈估計器去幫助我們建立整個異常辨識的準則，另外與其他用於異常檢測的方法相結合，其結果上也展示了這想法的可行性，另外也有與其他模型相比的結果。在實驗結果方面，在 MVTec- AD 資料集中有 15 類用於異常檢測的工具零件，平均的結果上，我們模型在 2、4 以及 8 張圖片訓練之下圖片異常分類和圖片異常分割之接收者操作特徵曲線底下面積(AUCROC)分數為 94.5(%)/ 96.9(%)、96.2(%)/ 97.6(%)、97.0(%)/ 97.8(%)。

In anomaly detection, the goal is to use the computer to distinguish between abnormal and normal objects to achieve automatic quality management. However, obtaining abnormal samples is difficult, making it challenging to collect both enough abnormal samples and all the types of the defects. As a result, anomaly detection models often rely on training with normal samples using unsupervised learning. This approach helps in learning and obtaining a dedicated model, which can predict whether a given testing sample is abnormal or not.
To train a model, a large number of images must typically be collected. However, the human eye can recognize anomaly images after seeing only a few normal images. So, this paper aims to achieve anomaly detection with a small number of normal samples for training.
Inspired by the normalizing flow, this paper proposes using a probability distribution estimator to establish the criterion for anomaly recognition. By combining this method with other anomaly detection techniques, the results demonstrate the feasibility of the approach. Finally, a comparison with other models will be presented.
Regarding the experimental results, the MVTec-AD datasets include 15 types of objects for anomaly detection. On average, our model achieves AUCROC scores of 94.5/96.9, 96.2/97.6, and 97.0/97.8 (%) for image anomaly classification/segmentation with 2, 4, and 8 training images, respectively.

[1] P. Bergmann, S. Löwe, M. Fauser, D. Sattlegger, and C. Steger, "Improving unsupervised defect segmentation by applying structural similarity to autoencoders," arXiv preprint arXiv:1807.02011, 2018.
[2] T. Matsubara, K. Sato, K. Hama, R. Tachibana, and K. Uehara, "Deep generative model using unregularized score for anomaly detection with heterogeneous complexity," IEEE Transactions on Cybernetics, vol. 52, no. 6, pp. 5161-5173, 2020.
[3] 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. 14902-14912.
[4] 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.
[5] T. Schlegl, P. Seeböck, S. M. Waldstein, U. Schmidt-Erfurth, and G. Langs, "Unsupervised anomaly detection with generative adversarial networks to guide marker discovery," in International conference on information processing in medical imaging, 2017: Springer, pp. 146-157.
[6] T. Schlegl, P. Seeböck, S. M. Waldstein, G. Langs, and U. Schmidt-Erfurth, "f-AnoGAN: Fast unsupervised anomaly detection with generative adversarial networks," Medical image analysis, vol. 54, pp. 30-44, 2019.
[7] J. Yu et al., "Fastflow: Unsupervised anomaly detection and localization via 2d normalizing flows," arXiv preprint arXiv:2111.07677, 2021.
[8] D. Gudovskiy, S. Ishizaka, and K. Kozuka, "Cflow-ad: Real-time unsupervised anomaly detection with localization via conditional normalizing flows," in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2022, pp. 98-107.
[9] N. Cohen and Y. Hoshen, "Sub-image anomaly detection with deep pyramid correspondences," arXiv preprint arXiv:2005.02357, 2020.
[10] T. Defard, A. Setkov, A. Loesch, and R. Audigier, "Padim: a patch distribution modeling framework for anomaly detection and localization," in International Conference on Pattern Recognition, 2021: Springer, pp. 475-489.
[11] K. Roth, L. Pemula, J. Zepeda, B. Schölkopf, 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. 14318-14328.
[12] 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.
[13] M. Rudolph, B. Wandt, and B. 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, 2021, pp. 1907-1916.
[14] A. Krizhevsky, I. Sutskever, and G. E. Hinton, "Imagenet classification with deep convolutional neural networks," Advances in neural information processing systems, vol. 25, 2012.
[15] E. Schwartz et al., "MAEDAY: MAE for few and zero shot AnomalY-Detection," arXiv preprint arXiv:2211.14307, 2022.
[16] K. He, X. Chen, S. Xie, Y. Li, P. Dollár, and R. Girshick, "Masked autoencoders are scalable vision learners," in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 16000-16009.
[17] S. Sheynin, S. Benaim, and L. Wolf, "A hierarchical transformation-discriminating generative model for few shot anomaly detection," in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 8495-8504.
[18] C. Huang, H. Guan, A. Jiang, Y. Zhang, M. Spratling, and Y.-F. Wang, "Registration based few-shot anomaly detection," in European Conference on Computer Vision, 2022: Springer, pp. 303-319.
[19] D. Rezende and S. Mohamed, "Variational inference with normalizing flows," in International conference on machine learning, 2015: PMLR, pp. 1530-1538.
[20] K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770-778.
[21] P. Mishra, R. Verk, D. Fornasier, C. Piciarelli, and G. L. Foresti, "VT-ADL: A vision transformer network for image anomaly detection and localization," in 2021 IEEE 30th International Symposium on Industrial Electronics (ISIE), 2021: IEEE, pp. 01-06.
[22] H. Touvron, M. Cord, M. Douze, F. Massa, A. Sablayrolles, and H. Jégou, "Training data-efficient image transformers & distillation through attention," in International conference on machine learning, 2021: PMLR, pp. 10347-10357.
[23] H. Touvron, M. Cord, A. Sablayrolles, G. Synnaeve, and H. Jégou, "Going deeper with image transformers," in Proceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 32-42.
[24] W. Liu, H. Chang, B. Ma, S. Shan, and X. Chen, "Diversity-measurable anomaly detection," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 12147-12156
[25] J. Shlens, "A tutorial on principal component analysis," arXiv preprint arXiv:1404.1100, 2014.
[26] P. Bergmann, M. Fauser, D. Sattlegger, and C. Steger, "Uninformed students: Student-teacher anomaly detection with discriminative latent embeddings," in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 4183-4192.
[27] S. Zagoruyko and N. Komodakis, "Wide residual networks," arXiv preprint arXiv:1605.07146, 2016.
[28] T. Cover and P. Hart, "Nearest neighbor pattern classification," IEEE transactions on information theory, vol. 13, no. 1, pp. 21-27, 1967.
[29] A. Radford et al., "Learning transferable visual models from natural language supervision," in International conference on machine learning, 2021: PMLR, pp. 8748-8763.
[30] G. Xie, J. Wang, J. Liu, F. Zheng, and Y. Jin, "Pushing the limits of fewshot anomaly detection in industry vision: Graphcore," arXiv preprint arXiv:2301.12082, 2023.
[31] J. Jeong, Y. Zou, T. Kim, D. Zhang, A. Ravichandran, and O. Dabeer, "Winclip: Zero-/few-shot anomaly classification and segmentation," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 19606-19616.
[32] A. Radford et al., "Learning transferable visual models from natural language supervision," in International conference on machine learning, 2021: PMLR, pp. 8748-8763.
[33] Z. Liu, Y. Zhou, Y. Xu, and Z. Wang, "Simplenet: A simple network for image anomaly detection and localization," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 20402-20411.
[34] O. Rippel, P. Mertens, and D. Merhof, "Modeling the distribution of normal data in pre-trained deep features for anomaly detection," in 2020 25th International Conference on Pattern Recognition (ICPR), 2021: IEEE, pp. 6726-6733.
[35] M. Tan and Q. Le, "Efficientnet: Rethinking model scaling for convolutional neural networks," in International conference on machine learning, 2019: PMLR, pp. 6105-6114.
[36] C. Schuhmann et al., "Laion-400m: Open dataset of clip-filtered 400 million image-text pairs," arXiv preprint arXiv:2111.02114, 2021.
[37] S. Liu et al., "Grounding dino: Marrying dino with grounded pre-training for open-set object detection," arXiv preprint arXiv:2303.05499, 2023.
[38] J. Hyun, S. Kim, G. Jeon, S. H. Kim, K. Bae, and B. J. Kang, "ReConPatch: Contrastive Patch Representation Learning for Industrial Anomaly Detection," arXiv preprint arXiv:2305.16713, 2023.
[39] A. Mousakhan, T. Brox, and J. Tayyub, "Anomaly Detection with Conditioned Denoising Diffusion Models," arXiv preprint arXiv:2305.15956, 2023.
[40] K. Han, Y. Wang, J. Guo, Y. Tang, and E. Wu, "Vision gnn: An image is worth graph of nodes," Advances in Neural Information Processing Systems, vol. 35, pp. 8291-8303, 2022.