[1] [1] Bergmann P, Jin X, Sattlegger D, et al. The MVTec 3D-AD dataset for unsupervised 3D anomaly detection and localization[J/OL]. https://arXiv.org/abs/2112.09045.

[3] [3] Kingma D P, Dhariwal P. Glow: generative flow with invertible 1x1 convolutions[J/OL]. https://arXiv.org/abs/1807.03039.

[5] [5] Wang G, Han S, Ding E, et al. Student-teacher feature pyramid matching for unsupervised anomaly detection[J]. arXiv, 2021: arXiv.210304257.

[7] [7] Zavrtanik V, Kristan M, SKOAJ D. Reconstruction by inpainting for visual anomaly detection[J]. Pattern Recognition, 2021, 112: 107706.

[8] [8] Yu J, Zheng Y, Wang X, et al. FastFlow: unsupervised anomaly detection and localization via 2D normalizing flows[J/OL] https://arXiv.org/abs/2111.07677v2.

[10] [10] Yoshuabengio L D. Nice: Non-linear independent components estimation[J]. arXiv e-prints, 2014: arXiv 1410.8516.

[12] [12] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.

[13] [13] Dinh L, Sohl-Dickstein J, Bengio S. Density estimation using Real NVP[C]//International Conference on Learning Representations.2016.

[14] [14] Bergmann P, Jin X, Sattlegger D, et al. The MVTec 3D-AD dataset for unsupervised 3D anomaly detection and localization[C]//Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, 2022: 202-213.

[15] [15] Hrwitz E, Hoshen Y. Back to the feature: classical 3D features are (almost) all you need for 3D anomaly detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2023: 2967-2976.