MIST: Multiple Instance Self-Training Framework for Video Anomaly Detection.
Jia-Chang FengFa-Ting HongWei-Shi ZhengPublished in: CoRR (2021)
Keyphrases
- anomaly detection
- multiple instance
- multiple instance learning
- intrusion detection
- intrusion detection system
- network intrusion detection
- semi supervised learning
- anomalous behavior
- detecting anomalies
- probabilistic model
- network traffic
- one class support vector machines
- real valued
- unsupervised learning
- video sequences
- unlabeled data
- multi modal
- hidden markov models
- concept learning
- high dimensional
- pairwise
- training set
- pattern recognition
- decision trees