A correlation graph approach for unsupervised manifold learning in image retrieval tasks.
Daniel Carlos Guimarães PedronetteRicardo da Silva TorresPublished in: Neurocomputing (2016)
Keyphrases
- manifold learning
- semi supervised
- neighborhood graph
- discriminant projection
- image retrieval
- low dimensional
- diffusion maps
- dimensionality reduction
- nonlinear dimensionality reduction
- high dimensional
- subspace learning
- unsupervised learning
- dimension reduction
- feature extraction
- random walk
- graph construction
- supervised learning
- feature space
- laplacian eigenmaps
- manifold structure
- high dimensional data
- labeled data
- geodesic distance
- graph laplacian
- graph embedding
- transfer learning
- sparse representation
- visual features
- pairwise
- head pose estimation
- training set
- multiscale
- similarity measure