Unsupervised Salient Object Segmentation Based on Kernel Density Estimation and Two-Phase Graph Cut.
Zhi LiuRan ShiLiquan ShenYinzhu XueKing Ngi NganZhaoyang ZhangPublished in: IEEE Trans. Multim. (2012)
Keyphrases
- graph cuts
- kernel density estimation
- object segmentation
- salient objects
- shape prior
- mean shift
- segmentation algorithm
- superpixels
- energy function
- image segmentation
- markov random field
- foreground objects
- saliency map
- segmentation method
- multi label
- unsupervised learning
- foreground and background
- pixel level
- object detection
- object tracking
- multi class
- computer vision
- image processing
- visual attention
- shape model
- machine learning
- image regions
- active contours