Discrepancy-Based Active Learning for Weakly Supervised Bleeding Segmentation in Wireless Capsule Endoscopy Images.
Fan BaiXiaohan XingYutian ShenHan MaMax Q.-H. MengPublished in: MICCAI (8) (2022)
Keyphrases
- ground truth
- test images
- weakly supervised
- active learning
- capsule endoscopy
- semantic segmentation
- segmentation algorithm
- superpixels
- segmentation method
- object class
- computer aided
- topic models
- bounding box
- level set
- object detection
- image analysis
- narrow band
- machine learning
- tumor detection
- multiple images
- image regions
- graph cuts
- image database
- higher order
- input image
- image data
- training set
- three dimensional
- image segmentation