Metric: Segmentation AUPRO (higher is better)
| # | Model↕ | Segmentation AUPRO▼ | Augmentations | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | CDO | 93.75 | Yes | Collaborative Discrepancy Optimization for Relia... | 2023-02-17 | Code |
| 2 | TransFusion | 0.983 | No | TransFusion -- A Transparency-Based Diffusion Mo... | 2023-11-16 | Code |
| 3 | Shape-Guided | 0.976 | Yes | - | - | Code |
| 4 | 3DSR | 0.972 | No | Cheating Depth: Enhancing 3D Surface Anomaly Det... | 2023-11-02 | Code |
| 5 | M3DM | 0.964 | Yes | Multimodal Industrial Anomaly Detection via Hybr... | 2023-03-01 | Code |
| 6 | Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (BTF) | 0.959 | Yes | Back to the Feature: Classical 3D Features are (... | 2022-03-10 | Code |
| 7 | TransFusion | 0.947 | No | TransFusion -- A Transparency-Based Diffusion Mo... | 2023-11-16 | Code |
| 8 | CPMF (2D+3D) | 0.9293 | Yes | Complementary Pseudo Multimodal Feature for Poin... | 2023-03-23 | Code |
| 9 | CPMF (3D) | 0.923 | Yes | Complementary Pseudo Multimodal Feature for Poin... | 2023-03-23 | Code |
| 10 | CPMF (2D) | 0.9145 | Yes | Complementary Pseudo Multimodal Feature for Poin... | 2023-03-23 | Code |
| 11 | Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (SIFT) | 0.91 | No | Back to the Feature: Classical 3D Features are (... | 2022-03-10 | Code |
| 12 | 3DSR (3D) | 0.907 | No | Cheating Depth: Enhancing 3D Surface Anomaly Det... | 2023-11-02 | Code |
| 13 | Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (HoG) | 0.771 | No | Back to the Feature: Classical 3D Features are (... | 2022-03-10 | Code |
| 14 | Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (Depth iNet) | 0.755 | Yes | Back to the Feature: Classical 3D Features are (... | 2022-03-10 | Code |
| 15 | Voxel GAN | 0.639 | No | The MVTec 3D-AD Dataset for Unsupervised 3D Anom... | 2021-12-16 | Code |
| 16 | Voxel AE | 0.564 | No | The MVTec 3D-AD Dataset for Unsupervised 3D Anom... | 2021-12-16 | Code |
| 17 | Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (NSA) | 0.5572 | No | Back to the Feature: Classical 3D Features are (... | 2022-03-10 | Code |
| 18 | Voxel VM | 0.471 | No | The MVTec 3D-AD Dataset for Unsupervised 3D Anom... | 2021-12-16 | Code |
| 19 | Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (RaW) | 0.442 | No | Back to the Feature: Classical 3D Features are (... | 2022-03-10 | Code |
| 20 | Depth VM | 0.374 | No | The MVTec 3D-AD Dataset for Unsupervised 3D Anom... | 2021-12-16 | Code |
| 21 | Depth AE | 0.203 | No | The MVTec 3D-AD Dataset for Unsupervised 3D Anom... | 2021-12-16 | Code |
| 22 | Depth GAN | 0.143 | No | The MVTec 3D-AD Dataset for Unsupervised 3D Anom... | 2021-12-16 | Code |