Anomaly Detection On Chuk Avenue

评估指标

AUC

RBDC

TBDC

评测结果

各个模型在此基准测试上的表现结果

				Paper Title	Repository
MULDE-object-centric-micro	94.3%	-	-	MULDE: Multiscale Log-Density Estimation via Denoising Score Matching for Video Anomaly Detection
AMSRC	93.8%	-	-	A Video Anomaly Detection Framework based on Appearance-Motion Semantics Representation Consistency	-
SSMTL++v1	93.7%	40.9	82.1	SSMTL++: Revisiting Self-Supervised Multi-Task Learning for Video Anomaly Detection	-
AI-VAD	93.7%	-	-	An Attribute-based Method for Video Anomaly Detection
Background-Agnostic Framework+SSMCTB	93.2%	66.04	65.12	Self-Supervised Masked Convolutional Transformer Block for Anomaly Detection
SSMTL+UBnormal	93.0%	61.10	61.38	UBnormal: New Benchmark for Supervised Open-Set Video Anomaly Detection
Background- Agnostic Framework+SSPCAB	92.9%	65.99	-	Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detection
VideoPatchCore	92.8%	-	-	VideoPatchCore: An Effective Method to Memorize Normality for Video Anomaly Detection
DMAD	92.8%	-	-	Diversity-Measurable Anomaly Detection
PGM	92.72%	60.18	72.09	Bounding Boxes and Probabilistic Graphical Models: Video Anomaly Detection Simplified
Background-Agnostic Framework	92.3%	65.05	66.85	A Background-Agnostic Framework with Adversarial Training for Abnormal Event Detection in Video
SSMTL++v2	91.6%	47.8	85.2	-	-
SSMTL	91.5%	57.00	58.30	Anomaly Detection in Video via Self-Supervised and Multi-Task Learning
SD-MAE	91.3%	46.77	66.58	Self-Distilled Masked Auto-Encoders are Efficient Video Anomaly Detectors
MAMA	91.2%	-	-	Making Anomalies More Anomalous: Video Anomaly Detection Using a Novel Generator and Destroyer	-
VALD-GAN	91.03	-	-	VALD-GAN: video anomaly detection using latent discriminator augmented GAN	-
Two-stream	90.8%	-	-	Context Recovery and Knowledge Retrieval: A Novel Two-Stream Framework for Video Anomaly Detection
AnomalyRuler	89.7%	-	-	Follow the Rules: Reasoning for Video Anomaly Detection with Large Language Models
Cloze Test	89.6%	-	-	Cloze Test Helps: Effective Video Anomaly Detection via Learning to Complete Video Events
Narrowed Normality Clusters	88.9%	-	-	Detecting abnormal events in video using Narrowed Normality Clusters	-

0 of 35 row(s) selected.