Semi-Overcomplete Convolutional Auto-Encoder Embedding as Shape Priors for Deep Vessel Segmentation.
A. SadikineBogdan BadicJean-Pierre TasuVincent NobletDimitris VisvikisPierre-Henri ConzePublished in: ICIP (2022)
Keyphrases
- vessel segmentation
- shape prior
- image segmentation
- sparse coding
- region growing
- graph cuts
- deformable models
- active contours
- level set
- level set based segmentation
- segmentation method
- shape model
- blood vessels
- segmentation algorithm
- variational segmentation
- sparse representation
- retinal images
- vessel enhancement
- bit rate
- natural images
- statistical shape
- energy function
- pose invariant
- shape variations
- active contour model
- shape representation
- prior knowledge
- graph cut segmentation
- statistical models
- medical images
- multiscale
- bayesian framework
- prior information
- user interaction
- neural network
- low contrast
- generative model
- markov random field
- image processing
- shape analysis
- energy functional
- pattern recognition
- video sequences
- feature extraction