HyperAI초신경

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

한국어

HyperAI초신경

Anomaly Detection On Mpdd

평가 지표

Detection AUROC

Segmentation AUROC

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Detection AUROC	Segmentation AUROC	Paper Title	Repository
GLAD	97.5	98.7	GLAD: Towards Better Reconstruction with Global and Local Adaptive Diffusion Models for Unsupervised Anomaly Detection
LeMO	87.4	97.8	Towards Total Online Unsupervised Anomaly Detection and Localization in Industrial Vision	-
AdaCLIP	82.5	96.1	AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detection
ADSPR	-	96.8	Anomaly Detection using Score-based Perturbation Resilience
PatchCore	82.12	95.66	Towards Total Recall in Industrial Anomaly Detection
ULSAD	95.73	97.45	Revisiting Deep Feature Reconstruction for Logical and Structural Industrial Anomaly Detection
DMDD	98.10	98.96	Dual-Modeling Decouple Distillation for Unsupervised Anomaly Detection	-
FastRecon	82.5	97.9	FastRecon: Few-shot Industrial Anomaly Detection via Fast Feature Reconstruction	-
DiffusionAD	96.2	98.5	DiffusionAD: Norm-guided One-step Denoising Diffusion for Anomaly Detection
POUTA	97.5	-	Produce Once, Utilize Twice for Anomaly Detection	-
GLASS	99.6	99.4	A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localization
Dinomaly	97.2	99.1	Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection
KAnoCLIP	77.8	98.3	KAnoCLIP: Zero-Shot Anomaly Detection through Knowledge-Driven Prompt Learning and Enhanced Cross-Modal Integration	-
RealNet	96.3	98.2	RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detection
PBAS	97.7	98.8	Progressive Boundary Guided Anomaly Synthesis for Industrial Anomaly Detection

0 of 15 row(s) selected.