Probabilistic relevance feedback approach for content-based image retrieval based on gaussian mixture models.
Apostolos MarakakisNikolaos GalatsanosAristidis LikasAndreas StafylopatisPublished in: IET Image Process. (2009)
Keyphrases
- gaussian mixture model
- relevance feedback
- mixture model
- image retrieval
- improve the retrieval accuracy
- probabilistic model
- speaker recognition
- information retrieval systems
- expectation maximization
- em algorithm
- generative model
- maximum likelihood
- feature vectors
- query expansion
- low level features
- density estimation
- information retrieval
- speaker identification
- multimedia
- variational bayes
- retrieval accuracy
- probability density
- active learning
- covariance matrices
- gaussian mixture
- image search
- probability density function
- image processing
- retrieval model
- image representation
- feature space