Two-Step Active Learning for Instance Segmentation with Uncertainty and Diversity Sampling.
Ke YuStephen AlbroGiulia DeSalvoSuraj KothawadeAbdullah RashwanSasan TavakkolKayhan BatmanghelichXiaoqi YinPublished in: CoRR (2023)
Keyphrases
- active learning
- random sampling
- image segmentation
- segmentation accuracy
- sampling strategies
- level set
- medical images
- unlabeled instances
- learning strategies
- data sets
- region growing
- segmentation method
- sample selection
- stratified sampling
- selective sampling
- segmentation algorithm
- semi supervised
- monte carlo
- transfer learning
- active sampling
- shape prior
- uncertain data
- batch mode
- sampling algorithm
- experimental design
- learning algorithm
- preprocessing step
- fully automatic
- medical imaging
- training examples
- post processing
- unlabeled data
- semi supervised learning
- image analysis
- multiscale
- image regions
- version space
- sampling strategy
- supervised learning
- pool based active learning