Weakly-Supervised Nucleus Segmentation Based on Point Annotations: A Coarse-to-Fine Self-Stimulated Learning Strategy.
Kuan TianJun ZhangHaocheng ShenKezhou YanPei DongJianhua YaoShannon ChePifu LuoXiao HanPublished in: MICCAI (5) (2020)
Keyphrases
- coarse to fine
- learning strategies
- weakly supervised
- multiscale
- superpixels
- deformable surface model
- object detection
- object class
- multiresolution
- object segmentation
- segmentation algorithm
- image segmentation
- online learning
- relation extraction
- image registration
- active learning
- object detectors
- topic models
- named entities
- segmentation method
- natural images
- level set
- dynamic programming
- semi supervised
- multi class
- shape prior
- energy function
- medical images
- multi view
- learning algorithm
- information extraction
- training data
- image processing