DSAL: Deeply Supervised Active Learning from Strong and Weak Labelers for Biomedical Image Segmentation.
Ziyuan ZhaoZeng ZengKaixin XuCen ChenCuntai GuanPublished in: CoRR (2021)
Keyphrases
- active learning
- image segmentation
- supervised learning
- semi supervised
- learning algorithm
- machine learning
- multiscale
- unsupervised learning
- supervised classification
- crowd sourced
- boundary detection
- semi supervised learning
- labeled data
- learning strategies
- level set method
- cost sensitive
- selective sampling
- experimental design
- graph cuts
- random sampling
- segmentation algorithm
- training examples
- markov random field
- image processing
- training set
- learning process
- gray level
- text mining
- transfer learning
- method for image segmentation
- contour detection
- information extraction
- segmentation method
- pool based active learning
- computer vision
- batch mode
- feature selection
- multiple instance learning
- fuzzy c means
- shape prior
- deformable models
- relevance feedback
- level set
- energy functional
- segmented images
- region merging
- biomedical images
- ground truth
- image segmentation algorithms
- object segmentation
- biomedical data
- pairwise
- unsupervised image segmentation