Anomaly Detection On Shanghaitech

评估指标

AUC

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
DAC(STG-NF + Jigsaw)	87.72%	Divide and Conquer in Video Anomaly Detection: A Comprehensive Review and New Approach
MULDE-object-centric-micro	86.7%	MULDE: Multiscale Log-Density Estimation via Denoising Score Matching for Video Anomaly Detection
AI-VAD	85.94%	An Attribute-based Method for Video Anomaly Detection
STG-NF	85.9%	Normalizing Flows for Human Pose Anomaly Detection
AnomalyRuler	85.2%	Follow the Rules: Reasoning for Video Anomaly Detection with Large Language Models
VideoPatchCore	85.1%	VideoPatchCore: An Effective Method to Memorize Normality for Video Anomaly Detection
Jigsaw-VAD	84.3%	Video Anomaly Detection by Solving Decoupled Spatio-Temporal Jigsaw Puzzles
SSMTL++v2	83.8%	SSMTL++: Revisiting Self-Supervised Multi-Task Learning for Video Anomaly Detection	-
SSMTL+UBnormal	83.7%	UBnormal: New Benchmark for Supervised Open-Set Video Anomaly Detection
two-stream	83.7%	Context Recovery and Knowledge Retrieval: A Novel Two-Stream Framework for Video Anomaly Detection
Background- Agnostic Framework+SSPCAB	83.6%	Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detection
SSMTL+++SSMCTB	83.6%	Self-Supervised Masked Convolutional Transformer Block for Anomaly Detection
MoPRL	83.35	Regularity Learning via Explicit Distribution Modeling for Skeletal Video Anomaly Detection
SSMTL++v1	82.9%	SSMTL++: Revisiting Self-Supervised Multi-Task Learning for Video Anomaly Detection	-
Background-Agnostic Framework	82.7%	A Background-Agnostic Framework with Adversarial Training for Abnormal Event Detection in Video
SSMTL	82.4%	Anomaly Detection in Video via Self-Supervised and Multi-Task Learning
MULDE-frame-centric-micro	81.3%	MULDE: Multiscale Log-Density Estimation via Denoising Score Matching for Video Anomaly Detection
TSGAD	80.6%	An Exploratory Study on Human-Centric Video Anomaly Detection through Variational Autoencoders and Trajectory Prediction
DMAD	78.8%	Diversity-Measurable Anomaly Detection
Object-centric AE	78.7%	Object-centric Auto-encoders and Dummy Anomalies for Abnormal Event Detection in Video

0 of 31 row(s) selected.