PEFAT: Boosting Semi-Supervised Medical Image Classification via Pseudo-Loss Estimation and Feature Adversarial Training.
Qingjie ZengYutong XieZilin LuYong XiaPublished in: CVPR (2023)
Keyphrases
- image classification
- semi supervised
- partially labeled data
- image features
- supervised learning
- medical students
- training process
- early stopping
- semi supervised learning
- image representation
- feature extraction
- batch mode
- spatial pooling
- spatial pyramid
- fully labeled
- multi view
- feature vectors
- training set
- boosting framework
- weak classifiers
- co training
- learning algorithm
- medical diagnosis
- pairwise
- unlabeled data
- machine learning
- labeled data
- cost sensitive
- class specific
- training samples
- spatial pyramid matching
- bag of words
- pairwise constraints
- semi supervised clustering
- training examples
- labelled data
- class labels
- feature set
- support vector machine
- training data
- visual words