Cross-Modal Disagreement-Guided Reliability-Aware Scoring for RGB-3D Industrial Anomaly Detection
Source abstract
Abstract
RGB–3D industrial anomaly detection seeks to jointly exploit texture and geometric cues for robust defect inspection. However, existing multimodal fusion methods still face two practical limitations: modality-specific anomaly evidence is often weakened after direct fusion, and image-level decisions remain unstable on difficult categories. To address these issues, this study develops a reliability-aware scoring enhancement on top of the released Hybrid Fusion/M3DM memory-bank pipeline. The method constructs a disagreement cue from RGB and point-cloud anomaly responses to enhance suspicious local regions and introduces a dual-branch image-level score calibration that combines a sensitive fusion branch with a robust statistical branch. Evaluated on MVTec 3D-AD under the official released-code full setting, the proposed method achieves 0.800 image-level ROCAUC, 0.980 pixel-level ROCAUC, and 0.926 AU-PRO, compared with 0.779, 0.975, and 0.915 for the corresponding released-code baseline in our environment. Additional evaluation on Eyecandies improves pixel-level ROCAUC and AU-PRO, while showing that image-level calibration remains dataset-sensitive. On a supplementary three-category Real-IAD D3 subset, the mean image-level ROCAUC, pixel-level ROCAUC, and AU-PRO improve from 0.963, 0.979, and 0.921 to 0.980, 0.988, and 0.941, respectively. These results indicate that explicit cross-modal disagreement modeling improves localization consistency, while image-level score calibration provides dataset-dependent gains rather than a uniform cross-dataset guarantee.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Anomaly Detection Techniques and Applications · Advanced Neural Network Applications · Domain Adaptation and Few-Shot Learning
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Shenyang Institute of Computing Technology, Chinese Academy of Sciences, Shenyang 110168, China
University of Chinese Academy of Sciences, Beijing 100049, China
Liaoning Province Human-Computer Interaction System Engineering Research Center Based on Digital Twin, Shenyang 110168, China
State Key Laboratory of Advanced Casting Technologies, Shenyang 110022, China
China Academy of Machinery Shenyang Research Institute of Foundry Co., Ltd., Shenyang 110022, China
References (30)
Show source-provided references
1. The MVTec Anomaly Detection Dataset: A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection10.1007/s11263-020-01400-4 ↗
2. A Survey on RGB, 3D, and Multimodal Approaches for Unsupervised Industrial Image Anomaly Detection10.1016/j.inffus.2025.103139 ↗
3. Industrial-Application-Oriented 2D Image and 3D Object Anomaly Detection Technology: A Comprehensive Review10.1007/s10489-025-06689-7 ↗
4. Roth, K., Pemula, L., Zepeda, J., Schölkopf, B., Brox, T., and Gehler, P. (2022, January 19–24). Towards Total Recall in Industrial Anomaly Detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA.10.1109/CVPR52688.2022.01392 ↗
5. Defard, T., Setkov, A., Loesch, A., and Audigier, R. (2021). PaDiM: A Patch Distribution Modeling Framework for Anomaly Detection and Localization. Pattern Recognition. ICPR International Workshops and Challenges, Springer.10.1007/978-3-030-68799-1_35 ↗
6. Bergmann, P., Jin, X., Sattlegger, D., and Steger, C. (2022, January 5–7). The MVTec 3D-AD Dataset for Unsupervised 3D Anomaly Detection and Localization. Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISAPP), Virtual.10.5220/0010865000003124 ↗
7. Wang, Y., Peng, J., Zhang, J., Yi, R., Wang, Y., and Wang, C. (2023, January 18–22). Multimodal Industrial Anomaly Detection via Hybrid Fusion. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada.10.1109/CVPR52729.2023.00776 ↗
8. Liu, J., Xie, G., Chen, R., Li, X., Wang, J., Liu, Y., Wang, C., and Zheng, F. (2023). Real3D-AD: A Dataset of Point Cloud Anomaly Detection. Advances in Neural Information Processing Systems 36 (NeurIPS 2023), Datasets and Benchmarks Track, Neural Information Processing Systems Foundation, Inc.. Available online: https://proceedings.neurips.cc/paper_files/paper/2023/hash/611b896d447df43c898062358df4c114-Abstract-Datasets_and_Benchmarks.html.
9. PointCore: An Efficient Framework for Unsupervised Point Cloud Anomaly Detection Using Joint Local–Global Features10.1016/j.neunet.2025.108446 ↗
10. Bergmann, P., Fauser, M., Sattlegger, D., and Steger, C. (2020, January 14–19). Uninformed Students: Student–Teacher Anomaly Detection with Discriminative Latent Embeddings. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.10.1109/CVPR42600.2020.00424 ↗
11. Zavrtanik, V., Kristan, M., and Skočaj, D. (2021, January 11–17). DRAEM: A Discriminatively Trained Reconstruction Embedding for Surface Anomaly Detection. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Montreal, QC, Canada.10.1109/ICCV48922.2021.00822 ↗
12. Deng, H., and Li, X. (2022, January 19–24). Anomaly Detection via Reverse Distillation from One-Class Embedding. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA.10.1109/CVPR52688.2022.00951 ↗
13. Li, C.-L., Sohn, K., Yoon, J., and Pfister, T. (2021, January 19–25). CutPaste: Self-Supervised Learning for Anomaly Detection and Localization. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Virtual.10.1109/CVPR46437.2021.00954 ↗
14. Liu, Z., Zhou, Y., Xu, Y., and Wang, Z. (2023, January 18–22). SimpleNet: A Simple Network for Image Anomaly Detection and Localization. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada.10.1109/CVPR52729.2023.01954 ↗
15. AEKD: Unsupervised Auto-Encoder Knowledge Distillation for Industrial Anomaly Detection10.1016/j.jmsy.2024.02.001 ↗
16. Horwitz, E., and Hoshen, Y. (2022). Back to the Feature: Classical 3D Features Are (Almost) All You Need for 3D Anomaly Detection. arXiv.10.1109/CVPRW59228.2023.00298 ↗
17. Wang, C., Zhu, W., Gao, B.-B., Gan, Z., Zhang, J., Gu, Z., Qian, S., Chen, M., and Ma, L. (2024, January 16–22). Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA. Available online: https://openaccess.thecvf.com/content/CVPR2024/html/Wang_Real-IAD_A_Real-World_Multi-View_Dataset_for_Benchmarking_Versatile_Industrial_Anomaly_CVPR_2024_paper.html.10.1109/CVPR52733.2024.02159 ↗
18. Zhu, W., Wang, L., Zhou, Z., Wang, C., Pan, Y., Zhang, R., Chen, Z., Cheng, L., Gao, B.-B., and Zhang, J. (2025). Real-IAD D3: A Real-World 2D/Pseudo-3D/3D Dataset for Industrial Anomaly Detection. arXiv.10.1109/CVPR52734.2025.01417 ↗
19. DAUP: Enhancing Point Cloud Homogeneity for 3D Industrial Anomaly Detection via Density-Aware Point Cloud Upsampling10.1016/j.aei.2024.102823 ↗
20. Fusion-Restoration Model for Industrial Multimodal Anomaly Detection10.1016/j.neucom.2025.130073 ↗
21. CPIR: Multimodal Industrial Anomaly Detection via Latent Bridged Cross-Modal Prediction and Intra-Modal Reconstruction10.1016/j.aei.2025.103240 ↗
22. M3DM-NR: RGB–3D Noisy-Resistant Industrial Anomaly Detection via Multimodal Denoising10.1109/TPAMI.2025.3592089 ↗
23. Li, Y., Yang, X., Zhang, J., Tian, S., Liao, J., and Liu, F. (2026, January 24–28). PIRN: Prototypical-Based Intra-Modal Reconstruction with Normality Communication for Multimodal Anomaly Detection. Proceedings of the International Conference on Learning Representations (ICLR), Rio de Janeiro, Brazil. Available online: https://openreview.net/forum?id=7L7kmHHfgf.
24. Costanzino, A., Zama Ramirez, P., Lisanti, G., and Di Stefano, L. (2024, January 16–22). Multimodal Industrial Anomaly Detection by Crossmodal Feature Mapping. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.10.1109/CVPR52733.2024.01631 ↗
25. Tao, C., Cao, X., and Du, J. (2025). G2SF-MIAD: Geometry-Guided Score Fusion for Multimodal Industrial Anomaly Detection. arXiv.
26. 3D Vision-Based Anomaly Detection in Manufacturing: A Survey10.1007/s42524-025-4189-9 ↗
27. Caron, M., Touvron, H., Misra, I., Jégou, H., Mairal, J., Bojanowski, P., and Joulin, A. (2021, January 11–17). Emerging Properties in Self-Supervised Vision Transformers. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Montreal, QC, Canada.10.1109/ICCV48922.2021.00951 ↗
28. Pang, Y., Wang, W., Tay, F.E.H., Liu, W., Tian, Y., and Yuan, L. (2022). Masked Autoencoders for Point Cloud Self-Supervised Learning. Computer Vision—ECCV 2022, Springer.10.1007/978-3-031-20086-1_35 ↗
29. Bonfiglioli, L., Toschi, M., Silvestri, D., Fioraio, N., and De Gregorio, D. (2023). The Eyecandies Dataset for Unsupervised Multimodal Anomaly Detection and Localization. Computer Vision—ACCV 2022, Springer.10.1007/978-3-031-26348-4_27 ↗
30. Li, Z., Kang, X., West, J., and Khoshelham, K. (2025). Out-of-Distribution Detection in 3D Applications: A Review. arXiv.10.2139/ssrn.5256108 ↗