Synthetic Pseudo Anomalies for Unsupervised Video Anomaly Detection: A Simple Yet Efficient Framework Based on Masked Autoencoder.
Xiangyu HuangCaidan ZhaoChenxing GaoLvdong ChenZhiqiang WuPublished in: ICASSP (2023)
Keyphrases
- anomaly detection
- unsupervised anomaly detection
- intrusion detection
- anomalous behavior
- detecting anomalies
- novelty detection
- unsupervised learning
- network intrusion detection
- network security
- behavior analysis
- network anomaly detection
- computer security
- intrusion detection system
- network traffic
- one class support vector machines
- detect anomalies
- video data
- data sets
- dimensionality reduction
- video sequences
- detecting anomalous
- data assimilation
- neural network
- pattern mining
- outlier detection
- model selection
- probabilistic model
- object recognition
- genetic algorithm
- negative selection algorithm