HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Anomaly Detection On Btad

평가 지표

Detection AUROC

Segmentation AUPRO

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Detection AUROC	Segmentation AUPRO	Paper Title	Repository
ReConPatch WRN-50	95.8	97.5	ReConPatch : Contrastive Patch Representation Learning for Industrial Anomaly Detection
PatchSVDD	-	-	Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentation
CPR	94.8	85.1	Target before Shooting: Accurate Anomaly Detection and Localization under One Millisecond via Cascade Patch Retrieval
MuSc (zero-shot)	96.16	83.43	MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled Images
RealNet	96.1	-	RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detection
FastFlow+AltUB	-	-	AltUB: Alternating Training Method to Update Base Distribution of Normalizing Flow for Anomaly Detection	-
PNI	-	-	PNI : Industrial Anomaly Detection using Position and Neighborhood Information
URD	93.9	78.5	Unlocking the Potential of Reverse Distillation for Anomaly Detection
AD-CLSCNFs	95.93	72.77	Anomaly Detection Using Normalizing Flow-Based Density Estimation and Synthetic Defect Classification
WeakREST-Un	94.4	84.9	Industrial Anomaly Detection and Localization Using Weakly-Supervised Residual Transformers	-
VT-ADL	-	-	VT-ADL: A Vision Transformer Network for Image Anomaly Detection and Localization
D3AD	95.2	83.2	Dynamic Addition of Noise in a Diffusion Model for Anomaly Detection
Reverse Distillation ++	95.63	-	Revisiting Reverse Distillation for Anomaly Detection
PyramidFlow (Res18)	95.8	-	PyramidFlow: High-Resolution Defect Contrastive Localization using Pyramid Normalizing Flow

0 of 14 row(s) selected.