Certifying Global Optimality of Graph Cuts via Semidefinite Relaxation: A Performance Guarantee for Spectral Clustering.
Shuyang LingThomas StrohmerPublished in: Found. Comput. Math. (2020)
Keyphrases
- spectral clustering
- graph cuts
- semidefinite
- convex relaxation
- globally optimal
- image segmentation
- multi label
- energy function
- pairwise
- markov random field
- energy minimization
- semidefinite programming
- data clustering
- shape prior
- belief propagation
- sufficient conditions
- higher dimensional
- k means
- interior point methods
- clustering algorithm
- image processing
- convex sets
- finite dimensional
- convex optimization
- unsupervised learning