Fast likelihood computation for continuous-mixture densities using a tree-based nearest neighbor search.
Frank SeidePublished in: EUROSPEECH (1995)
Keyphrases
- nearest neighbor search
- gaussian densities
- kd tree
- nearest neighbor
- sequential search
- high dimensional
- image retrieval
- memory based learning
- feature space
- search methods
- fractal image compression
- similarity search
- locality sensitive hashing
- cost model
- exhaustive search
- kd trees
- index structure
- manhattan distance
- maximum likelihood
- approximate nearest neighbor
- mixture of gaussians
- knn
- learning algorithm
- face recognition
- feature extraction
- data structure
- road network
- neural network
- expectation maximization