Deep Anomaly Discovery from Unlabeled Videos via Normality Advantage and Self-Paced Refinement.
Guang YuSiqi WangZhiping CaiXinwang LiuChuanfu XuChengkun WuPublished in: CVPR (2022)
Keyphrases
- anomaly detection
- video sequences
- unsupervised learning
- video database
- active learning
- unlabeled data
- knowledge discovery
- scientific discovery
- intrusion detection
- video content
- emotional intelligence
- discovery process
- deep learning
- video frames
- video data
- labeled data
- data points
- data mining
- abnormal events
- refinement process
- detect anomalies
- positive examples
- brain computer interface
- human actions
- class labels
- learning algorithm