Leveraging triplet loss for unsupervised action segmentation.
Elena Belén Bueno-BenitoBiel Tura VecinoMariella DimiccoliPublished in: CoRR (2023)
Keyphrases
- image segmentation
- fully unsupervised
- text segmentation
- bayesian image segmentation
- segmentation method
- segmentation algorithm
- brain mri
- segmentation accuracy
- semi supervised
- shape prior
- region growing
- segmentation errors
- multiscale
- unsupervised learning
- level set
- medical images
- markov fields
- unsupervised clustering
- segmentation of textured images
- computer vision
- normalized cut
- pixel level
- fully automatic
- ground truth
- edge detection
- texture segmentation
- background subtraction
- contour detection
- image analysis
- shape model
- optimal segmentation
- model selection
- supervised learning