HyperAI超神経

プラットフォーム

Rgb 3D Anomaly Detection And Segmentation On

評価指標

Detection AUCROC

Segmentation AUPRO

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

			Paper Title
TransFusion	0.982	0.983	TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection
3DSR	0.978	0.972	Cheating Depth: Enhancing 3D Surface Anomaly Detection via Depth Simulation
Shape-Guided	0.947	0.976	Shape-Guided: Shape-Guided Dual-Memory Learning for 3D Anomaly Detection
M3DM	0.945	0.964	Multimodal Industrial Anomaly Detection via Hybrid Fusion
AST	0.937	-	Asymmetric Student-Teacher Networks for Industrial Anomaly Detection
Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (BTF)	0.865	0.959	Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection
Voxel VM	0.609	0.471	The MVTec 3D-AD Dataset for Unsupervised 3D Anomaly Detection and Localization
Voxel AE	0.538	0.564	The MVTec 3D-AD Dataset for Unsupervised 3D Anomaly Detection and Localization
Voxel GAN	0.517	0.639	The MVTec 3D-AD Dataset for Unsupervised 3D Anomaly Detection and Localization

0 of 9 row(s) selected.