Training-Based Multiple Source Tracking Using Manifold-Learning and Recursive Expectation-Maximization.
Avital BrossSharon GannotPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2023)
Keyphrases
- manifold learning
- multiple sources
- expectation maximization
- semi supervised
- domain adaptation
- manifold embedding
- low dimensional
- em algorithm
- dimensionality reduction
- diffusion maps
- nonlinear dimensionality reduction
- high dimensional
- high dimensional data
- feature extraction
- head pose estimation
- dimension reduction
- particle filter
- training set
- data sets
- training samples
- locally linear embedding
- visual tracking
- computer vision
- machine learning
- riemannian manifolds
- k means
- target domain
- gaussian mixture
- pattern recognition
- semi supervised learning
- maximum likelihood
- multi view
- supervised learning