Nonlinear Dimensionality Reduction by Unit Ball Embedding (UBE) and Its Application to Image Clustering.
Behrouz Haji SoleimaniStan MatwinPublished in: ICMLA (2016)
Keyphrases
- nonlinear dimensionality reduction
- image data
- high dimensional data
- input image
- manifold learning
- image representation
- image classification
- clustering algorithm
- image features
- dimensionality reduction
- image segmentation
- maximum variance unfolding
- clustering method
- distance transform
- data sets
- laplacian eigenmaps
- nearest neighbor
- semi supervised
- similarity measure
- high dimensionality
- intensity images