Segmentation in Style: Unsupervised Semantic Image Segmentation with Stylegan and CLIP.
Daniil PakhomovSanchit HiraNarayani WagleKemar E. GreenNassir NavabPublished in: CoRR (2021)
Keyphrases
- image segmentation
- segmentation algorithm
- segmentation method
- shape prior
- multiscale
- region growing
- texture image segmentation
- otsu method
- normalized cut
- level set
- image segmentation algorithms
- unsupervised image segmentation
- joint segmentation
- unsupervised clustering
- unsupervised segmentation
- object segmentation
- region merging
- segmented images
- medical imaging
- texture segmentation
- region segmentation
- contour detection
- segmentation procedure
- segmenting images
- image analysis
- fuzzy connectedness
- graph cuts
- level set method
- image segmentation algorithm
- fully unsupervised
- deformable models
- berkeley segmentation dataset
- active contours
- boundary information
- segmentation result
- computer vision
- energy functional
- energy function
- markov random field
- final segmentation
- markov random field model
- watershed transformation
- graph based image segmentation
- text segmentation
- graph partitioning
- image processing
- segmented regions
- boundary detection
- interactive segmentation
- image segmentation methods
- watershed segmentation
- model based segmentation
- spectral clustering
- semantic information
- colour image segmentation
- variational framework
- perceptual grouping
- adaptive thresholding
- video clips
- low level features
- gray level
- high level