Multiple Feature Fusion Based on Co-Training Approach and Time Regularization for Place Classification in Wearable Video.
Vladislavs DovgalecsRémi MégretYannick BerthoumieuPublished in: Adv. Multim. (2013)
Keyphrases
- co training
- text classification
- multiple feature fusion
- semi supervised learning
- multi view
- semi supervised
- supervised learning
- unlabeled data
- multi view learning
- single view
- unlabelled data
- video sequences
- semi supervised classification
- pattern recognition
- partially labeled data
- labeled and unlabeled data
- training examples
- learning algorithm
- support vector machine
- training set
- training process
- decision trees
- feature extraction
- support vector
- labeled data
- feature vectors
- named entities
- active learning
- classification accuracy
- multiple views
- cross validation
- data analysis
- class labels
- three dimensional
- email classification
- neural network