TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

SotA/Methodology/Anomaly Detection/MVTEC 3D-AD

Anomaly Detection on MVTEC 3D-AD

Metric: Detection AUROC (higher is better)

LeaderboardDataset
Loading chart...

Results

Submit a result
#Model↕Detection AUROC▼AugmentationsPaperDate↕Code
1AST93.7YesAsymmetric Student-Teacher Networks for Industri...2022-10-14Code
2TransFusion0.957NoTransFusion -- A Transparency-Based Diffusion Mo...2023-11-16Code
3MC4AD0.954YesExamining the Source of Defects from a Mechanica...2025-05-09Code
4CPMF (2D+3D)0.9515YesComplementary Pseudo Multimodal Feature for Poin...2023-03-23Code
5CPMF (2D+3D)0.9515YesComplementary Pseudo Multimodal Feature for Poin...2023-03-23Code
6GLFM0.944No--Code
73DSR (3D)0.922NoCheating Depth: Enhancing 3D Surface Anomaly Det...2023-11-02Code
8Shape-Guided (only SDF)0.916No--Code
9CPMF (2D)0.8918YesComplementary Pseudo Multimodal Feature for Poin...2023-03-23Code
10CPMF (2D)0.8918YesComplementary Pseudo Multimodal Feature for Poin...2023-03-23Code
11CPMF (3D)0.8304YesComplementary Pseudo Multimodal Feature for Poin...2023-03-23Code
12CPMF (3D)0.8304YesComplementary Pseudo Multimodal Feature for Poin...2023-03-23Code
13Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (FPFH)0.782NoBack to the Feature: Classical 3D Features are (...2022-03-10Code
14Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (SIFT)0.727NoBack to the Feature: Classical 3D Features are (...2022-03-10Code
15Voxel AE0.699NoThe MVTec 3D-AD Dataset for Unsupervised 3D Anom...2021-12-16Code
16Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (NSA)0.696NoBack to the Feature: Classical 3D Features are (...2022-03-10Code
17Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (Depth iNet)0.675YesBack to the Feature: Classical 3D Features are (...2022-03-10Code
18Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (RaW)0.573NoBack to the Feature: Classical 3D Features are (...2022-03-10Code
19Voxel VM0.571NoThe MVTec 3D-AD Dataset for Unsupervised 3D Anom...2021-12-16Code
20Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (HoG)0.559NoBack to the Feature: Classical 3D Features are (...2022-03-10Code
21Depth VM0.546NoThe MVTec 3D-AD Dataset for Unsupervised 3D Anom...2021-12-16Code
22Depth AE0.546NoThe MVTec 3D-AD Dataset for Unsupervised 3D Anom...2021-12-16Code
23Voxel GAN0.537NoThe MVTec 3D-AD Dataset for Unsupervised 3D Anom...2021-12-16Code
24Depth GAN0.523NoThe MVTec 3D-AD Dataset for Unsupervised 3D Anom...2021-12-16Code