Open-set semi-supervised audio-visual speaker recognition using co-training LDA and Sparse Representation Classifiers.
Xuran ZhaoNicholas W. D. EvansJean-Luc DugelayPublished in: ICASSP (2013)
Keyphrases
- co training
- audio visual
- sparse representation
- semi supervised
- face recognition
- unlabeled data
- semi supervised learning
- multi modal
- dimensionality reduction
- generative model
- linear discriminant analysis
- labeled data
- visual information
- multi view
- high dimensional data
- supervised learning
- visual data
- text classification
- image classification
- active learning
- pairwise
- topic models
- single view
- feature extraction
- subspace learning
- labeled and unlabeled data
- image representation
- semi supervised learning methods
- unsupervised learning
- training examples
- named entities
- face images
- nearest neighbor
- signal processing
- principal component analysis
- feature space
- gaussian process
- object recognition
- data sets
- machine learning
- learning algorithm
- video sequences
- pattern recognition
- data points
- small number
- facial expressions
- co occurrence