Topology-preserving transfer learning for weakly-supervised anomaly detection and segmentation.
Shenxing WeiXing WeiMuhammad Rifki KurniawanZhiheng MaYihong GongPublished in: Pattern Recognit. Lett. (2023)
Keyphrases
- anomaly detection
- transfer learning
- weakly supervised
- superpixels
- segmentation algorithm
- labeled data
- semi supervised
- segmentation method
- machine learning
- reinforcement learning
- topic models
- relation extraction
- image segmentation
- semi supervised learning
- text classification
- multiscale
- level set
- unsupervised learning
- active learning
- shape prior
- learning algorithm
- high dimensional
- named entities
- medical images
- data sets
- self organizing maps
- training data
- text categorization
- unlabeled data
- dimensionality reduction
- co occurrence
- prior knowledge
- knowledge discovery