HyperAI超神经

首页资讯最新论文教程数据集百科 SOTA LLM 模型天梯 GPU 天梯顶会

中文

HyperAI超神经

Depth Anomaly Detection And Segmentation On

评估指标

Detection AUROC

Segmentation AUPRO

Segmentation AUROC

评测结果

各个模型在此基准测试上的表现结果

模型名称	Detection AUROC	Segmentation AUPRO	Segmentation AUROC	Paper Title	Repository
CPMF (2D)	0.8918	0.9145	0.9730	Complementary Pseudo Multimodal Feature for Point Cloud Anomaly Detection	-
TransFusion	0.957	0.947	-	TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection	-
Depth VM	0.546	0.374	-	The MVTec 3D-AD Dataset for Unsupervised 3D Anomaly Detection and Localization	-
CPMF (2D+3D)	0.9515	0.9293	0.9781	Complementary Pseudo Multimodal Feature for Point Cloud Anomaly Detection	-
CPMF (3D)	0.8304	0.9230	0.9780	Complementary Pseudo Multimodal Feature for Point Cloud Anomaly Detection	-
Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (Depth iNet)	0.675	0.755	0.930	Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection	-
Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (SIFT)	0.727	0.910	0.974	Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection	-
Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (RaW)	0.573	0.442	0.771	Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection	-
3DSR (3D)	0.922	0.907	-	Cheating Depth: Enhancing 3D Surface Anomaly Detection via Depth Simulation	-
Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (NSA)	0.696	0.5572	0.817	Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection	-
Depth GAN	0.523	0.143	-	The MVTec 3D-AD Dataset for Unsupervised 3D Anomaly Detection and Localization	-
Depth AE	0.546	0.203	-	The MVTec 3D-AD Dataset for Unsupervised 3D Anomaly Detection and Localization	-
Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (HoG)	0.559	0.771	0.930	Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection	-

0 of 13 row(s) selected.

Depth Anomaly Detection And Segmentation On | SOTA | HyperAI超神经