VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection.
Peng WuXuerong ZhouGuansong PangLingru ZhouQingsen YanPeng WangYanning ZhangPublished in: CoRR (2023)
Keyphrases
- anomaly detection
- language model
- weakly supervised
- probabilistic model
- topic models
- n gram
- superpixels
- relation extraction
- video sequences
- information retrieval
- computer vision
- named entities
- video data
- semi supervised
- image processing
- unsupervised learning
- domain specific
- feature space
- pattern recognition
- knn
- knowledge discovery
- high dimensional
- pairwise
- named entity recognition
- object detectors