Explicit Topological Priors for Deep-Learning Based Image Segmentation Using Persistent Homology.
James R. CloughIlkay ÖksüzNicholas ByrneJulia A. SchnabelAndrew P. KingPublished in: IPMI (2019)
Keyphrases
- persistent homology
- deep learning
- image segmentation
- morse theory
- topological features
- unsupervised learning
- computational geometry
- machine learning
- unsupervised feature learning
- weakly supervised
- multiscale
- graph cuts
- mental models
- segmentation method
- computer vision
- shape prior
- deformable models
- active contours
- markov random field
- image processing
- segmentation algorithm
- semi supervised
- link prediction
- level set
- object recognition